[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/diy/ - Do It Yourself


View post   

File: 67 KB, 473x673, 1547871792232.jpg [View same] [iqdb] [saucenao] [google]
1737362 No.1737362 [Reply] [Original]

flew-the-coop edition: >>1731520

>RULES
0. Electrics ≠ electronics. Appliances/mains/sparky stuff to /qtddtot/ or /sqt/. PC assembly >>>/g/
1. Search web first. Re-read all documentation/data-sheets related to your components/circuits. THEN ask. Show your work.
2. Pics > 1000 words. Post relevant schematic/picture/sketch with all part numbers/values/etc when asking for help. Focus/lighting counts.
2.5. State your skill level if asking an open-ended question.
3. Read posts fully. Solve more problems than you create.
4. /ohm/ is an anonymous, non-smoking general.

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

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

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

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

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

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

>Li+/LiPo batteries
Read this first: http://www.elteconline.com/download/pdf/SAFT-RIC-LI-ION-Safety-Recommendations.pdf
>headphone jack noise
Look up "ground loop isolator".
>I have junk, what do?
Get rid of it.

>> No.1737382

So according to the pdf, to charge a lipo with a bench power supply, where the lipo is 1s 1AH, i set the power supply voltage to 4.2V and at the same time maximum current to 1A, is this correct?

>> No.1737416

>>1737292
>I just poped the back of the TV and 5 electrolytic capacitors are bulged.
What a story, Mark! Haha.
>>1737365
or you could just get brand-name shit from a reputable online seller and follow their datasheets...

I mean what's the actual cost difference gonna be?

>> No.1737417

>>1737382
you should really use a cutoff circuit that terminates the charge once the termination voltage has been reached. Li+ doesn't much like being trickle-charged

>> No.1737419

>>1737365
>trusting alibay
holy kek
you don't need to measure anything, just choose the appropriate series from the manufacturer's catalog

>> No.1737420

>>1737416
>>1737419
These. I don't know why you'd buy a handful of jellybean parts of aliexpress. Given you'll have to wait at least a month to find out if you got chinked (e.g., working pulls, dry caps, or otherwise rejected) for a savings of what? A couple bucks?

Fuck even if you don't care about getting high-end caps get them from somewhere reputable that has datasheets.

>> No.1737424

>>1737416
>>1737419
Dont be snarky guys I barely know what im doing and its not about how much money I need to spend on capacitors.
They can be $5 a piece as long as the problem is solved and the tv turns on.
I will pick one of the recommended sites in the op and order from there,RS Components seems nice.

>> No.1737427
File: 33 KB, 600x600, 13853-02.jpg [View same] [iqdb] [saucenao] [google]
1737427

>>1737382
>So according to the pdf
[citation needed]

>> No.1737428

>>1737424
Didn't think any of those posts were snarky. It's just a bad idea for a small project like that to have to wait weeks or months for jellybean parts.

>> No.1737429

>>1737428
For the waiting yeah I figured that myself but I don't have much of a choice considering no shop in town has them and the other solution is to search for a repair man (which I will do if it turns out its a bigger problem).
The only other thing is to order from a site I have experience ordering from,since I order shit from ali all the time I figured hey maybe their caps are not bad.
Now that I know thats not the case I will shop from a more reputable places for electronic components.

>> No.1737433

>>1737429
Amazon, regular eBay, digikey, any other online electronics supply houses? It's just... fuck man the stuff on aliexpress is so hit-and-miss.

>> No.1737472

>>1737433
mouser digikey and arrow are your best bets for authorized distributors where you know what your buying is legit.
LCSC is china and not "authorized" but so far I've never had a bad part from them plus they're orders cheaper than the above. Some of the brand IC's like TI parts have Chinese specific part numbers at the end but they function as the normal part. I would use LCSC as my cheapest option.
All others like amazon, eBay, aliexpress etc are crapshoots since it's 3rd party sellers on those marketplaces and depending on who you buy you can get fakes, wrong part numbers, etc so I'd avoid for IC's and stuff and passives like resistors are so cheap might as well get them from the above listed anyways.

>> No.1737475

>>1737472
>it's 3rd party sellers on those marketplaces and depending on who you buy you can get fakes, wrong part numbers, etc
Yeah I'm just saying that if you're gonna take that crapshoot on ICs and passives, at least Amazon and eBay (if you pick the right sellers) won't take a month and a half for you to find out if you got screwed.

Like there's absolutely nothing wrong with taking a crapshoot now and again if it's no big delay or frustration if it winds up not working out.

>> No.1737533
File: 1.03 MB, 748x477, dep.png [View same] [iqdb] [saucenao] [google]
1737533

Anon, did you ever use depletion-mode MOSFETs?

>> No.1737586
File: 7 KB, 126x174, AP7362.png [View same] [iqdb] [saucenao] [google]
1737586

This thread is brought to us by the AP7362 ultra-low-dropout linear regulator.
1.5 amps in a 2x3mm package!

>> No.1737598

>>1737533
no, are there any models commonly available?

>> No.1737611

Hello, I'm lernding but I have some confusion

Current leads voltage in a capacitor, I've read this over and over. Is this

i) an application of capacitors, and if so why would I want to do this

or

ii) an undesired side effect of capacitors, if so why, and if so how would I correct for it, "put in an inductor as well" seems like too straightforward an answer

>> No.1737612

>>1737611
>an application of capacitors
well sure, absolutely.
Try reading about "phase shift" and then "phase shift oscillator". that will help you understand it.
It's an undesired side effect when a discharged capacitor draws more current from a power rail than you would expect.

>> No.1737618

>>1737611
It's a result of the ideal characteristic of a capacitor: I = C*dV/dt. In addition to the simple phase shifting as pointed out by >>1737612, this can also be used to make integrators and differentiators.
But in some cases (e.g. filters in audio circuits) a phase shift can cause distortion, so there are certainly cases where there is a need to remove this phase change. But I believe that active (op amp or transistor) circuits are used instead of inductors, just like with signal filters in general. Inductors are really only used instead of op-amp virtual inductors for higher-power situations.

>> No.1737619

>>1737611
>>1737618
Oh and in addition an inductor is represented: V = L*dI/dt

>> No.1737628

>>1737611
iii) it's the whole point of a capacitor

>>1737586
noice

>>1737433
>implying Western (((marketplaces))) are as good as distributors, or even significantly better than aliexpress

>> No.1737637

>>1737628
>implying Western (((marketplaces))) are as good as distributors, or even significantly better than aliexpress
In terms of (1) getting your product in a timely fashion and (2) being able to resolve disputes, then yes.

Even if the counterfeit/fraud rate was much higher than Ali, for the application that guy was talking about, I'd still go with someone based outside of China.

>> No.1737649
File: 249 KB, 2002x1001, 1553155236706.jpg [View same] [iqdb] [saucenao] [google]
1737649

I got one of these remote switches for Christmas lights. Of course, it came with no instructions. Do I connect to NC (normally closed) or NO (normally open)? Anyone used one of these chink fire hazards?

>> No.1737650

>>1737649
probably NO

>> No.1737652

>>1737628
>he doesn't get free samples from TI

>> No.1737658

>>1737649
well do you want it to be normally open or normally closed

normally open means off until you activate the switch, turning it on

normally closed means on until you activate the switch, turning it off

link me the switches pls

>> No.1737660

>>1737649
did the one you got come with the shit stains on the bottom too?

>> No.1737666

>>1737660
Those aren't shitstains, just a bit of mess with the no-clean, that's all. Or with the conformal coating, butI don't think they have one.
That +V solder joint is a bit how-ya-doin though.

>> No.1737667

>>1737650
that's what I assumed as well. however, I just realized that I don't have a battery for the keyfob so can't test it until tomorrow.
>>1737658
>link me the switches pls
"12v remote switch" on your fav chinkshit merchant site
>>1737660
no. it was very clean and came in a sealed antistatic bag too.

>> No.1737743
File: 29 KB, 500x500, 41TgVE+vedL.jpg [View same] [iqdb] [saucenao] [google]
1737743

Hi /ohm/
So im trying to interface a load cell like pic related with a lab amplifier
Usually thoose load cell are used with this chineese board (HX711) i can't use it here because of the shitty sample rate of theese circuit (80SPS)
I want to use this cell in a PID regulation loop so this is much too low, i need to use the cell coupled to an amp board and an arduino ( i will be using a arduino DUE which ADC sample rate is 100 kS per second)

>> No.1737744
File: 75 KB, 617x322, loadcell.png [View same] [iqdb] [saucenao] [google]
1737744

>>1737743
I have some trouble with the lab amp im using :

All wiring schematic i've seen for this application seem to use a amplifier configuration where there is 2 input separate from ground (here V+IN and V-IN)

But my amp is not like that, the input is only one pin, the other "input" pin is the ground, same as output, continuity test show connection between input and ouput ground

>> No.1737745
File: 1.85 MB, 2736x3648, P1050533.jpg [View same] [iqdb] [saucenao] [google]
1737745

>>1737744
Heres my setup

>> No.1737746

>>1737743
>arduino
there's your problem

>> No.1737747
File: 500 KB, 2804x1236, P1050535.jpg [View same] [iqdb] [saucenao] [google]
1737747

>>1737745
i tryed this wiring, output show no response to load cell force, all i read is noise from the power supply

Im lost here, maybe my lab amp is not suitable for this work, is that Case i will make a custom one, but if it is possible to avoid it that would be nice !

>> No.1737751

>>1737744
Then you should have bought a damn instrumentation amplifier then, right? Could still get away with a floating power supply though, like with a 9V battery or whatever.

>>1737747
Check that the PSU output is floating, because if not you'll just be making a ground loop.
Also:
>usb oscilloscope
>not second-hand soviet crt scope

>> No.1737829
File: 43 KB, 700x393, file.png [View same] [iqdb] [saucenao] [google]
1737829

Okay, new plan for the simple digital controlled power supply

I will replace R3 and CC pots with a digital pots and boom done, that was the the pots should be hidden from any large voltages and stay well within their 5V spec

And please don't tell me it won't work, I am not redoing the whole thing fucking AGAIN

>> No.1737838

>>1737746
first of many

>>1737829
it might work
but digipots aren't for power handling at all. maybe spend the extra 2 kopek for a proper dropper resistor for HL2 and a separate diode to slam DA2's feedback input

>> No.1737844

>>1737829

the kangaroo-fucker on eevblog did a whole series of videos on building your own supply. he added digital control at the end.
(however he, inexplicably, only designed the thing to go 0-5.5V or so, making it substantially useless. my guess is he was paid off by Keysight, or whoever, to make a crippled version, so as not to compete with commercial units)

>> No.1737845
File: 17 KB, 600x201, capacitor bank.png [View same] [iqdb] [saucenao] [google]
1737845

>>1737829
Reee I said it in the last thread. You can generate a analog output with the arduino pwm.
>measure current
>compare current
>send pwm output to the current control circuit

And R3 is used to measure the output voltage, not the current. Usually converters have a sense resistor somewhere for this. (If you know your power electronics you can control current with a single feedback pin, but I did not read the datasheet for that IC so).

Digital Pots are not good for this! Get your basics straight and stop overcomplicating things with stuff like stepper controlled pots and so on. Just delete the capacitor bank part of the pic. Some dude mentioned dave's power supply video and what he said also applies!

>> No.1737849
File: 32 KB, 723x592, current sense1.png [View same] [iqdb] [saucenao] [google]
1737849

>>1737845
This current sensor works with most shitty opamps. You'll have to do the math to find the gain (I don't remember it now but it's very easy). In the pic the 10Ohm resistor is the load. If you google "current sense" you'll find a lot of diferential amps and instrumentation amps. They are ideal for this, but if you try making one with discrete shitty opamps it usually don't work because of common voltage range issues.

>> No.1737851

>>1737849
The output is the emitter resistors btw

>> No.1737858

So there's this dumbass who keeps ripping off branches from my bush that I keep in the underground garage over the winter.
What's an analog sensor that I can use that triggers on strong vibrations?
My plan is to trigger a horn hooked up to a motorcycle battery hidden behind the vase that should scare the crap out of him.

>> No.1737859

>>1737845
I don't understand what you mean. PWM is a square wave, you can't use that to replace the resistor, it would have to be flattened by something like a pass filter which gives you a wave not a perfect analog voltage anyway.
Because i already have digi pots i will try it that way because it's so super simple.
If i destroy the buck it's fine it only costs $2 and i have two of them
>>1737844
I watched that aussie tutorial on a bench supply and i liked it, except he used lm317s, so that is a lot of waste heat to deal with, but thats fane i got heatsinks for days, BUT for some reason, as you also mention, his design was 0-1V and up to 1A output. Fucking useless. What sort of fucking bench power supply can do 0 - 1V range? that won't even turn on a god damn led. The alternative chips he suggested in the video are like $6 a pop and at that point i can just buy a finished chink benchtop supply.

>> No.1737862

>>1737858
i did something vaguely similar just recently. get an MPU6150 or MPU9150 accelerometer. connect to an arduino nano. there's a library that makes communicating with those chips dead simple. you just need to measure how much the accelerometer readings deviate from their baseline in any given sample and then compare that to a threshold to see if you're vibrating. your challenge will be setting the threshold just right so you don't trigger on random vibrations like someone walking by.

>> No.1737872
File: 64 KB, 960x960, aura.jpg [View same] [iqdb] [saucenao] [google]
1737872

>>1737859
> it would have to be flattened by something like a pass filter which gives you a wave not a perfect analog voltage anyway.
it would have to be flattened by something like a pass filter which gives you a wave not a perfect analog voltage anyway.
> it would have to be flattened by something like a pass filter which gives you a wave not a perfect analog voltage anyway.
>perfect analog voltage anyway.
>reeee my 5v chip is receiving 4.92V! IT'S BROKEN DAD!!
>t. arduino general

>> No.1737876
File: 111 KB, 523x601, free-fall sensor.jpg [View same] [iqdb] [saucenao] [google]
1737876

>>1737858

if you have an old dead hard disk, it'll likely have a free-fall sensor (or 2) that can detect movement. find datasheet and apply.
(it's the white rectangle in center of pic)

>> No.1737877

>>1737862
I was hoping not having to use any MCUs
Though the accelerometer idea is not bad,
I might use an analog ADXL one and trim it with a comparator maybe.

>> No.1737886

So I have 2 identical cordless power tools with brushed motors.
Both brand new.

One motor has a higher whinier pitch than the other. They both seem to perform the same.

Is the sound signature indicative of anything, like quality control of the motor? Is one going to last longer in theory than the other?

>> No.1737888

>>1737877
scratch that
https://www.mouser.de/datasheet/2/60/rb-1280353.pdf
this should be easier

>> No.1737889

>>1737886
this can be caused by a ton of different things from harmonics in the bldc driver exciting the motor windings, different bearings/bushings, or different housing designs. it's not necessarily indicative of lower quality but in a very general sense less noise means less friction/vibration and thus less wear on parts. t. mechanical engineer

>> No.1737890

>>1737889
scratch the bldc drivers, i read at a first grade level and didn't notice the brushed motor part.

>> No.1737892
File: 795 KB, 1000x1333, howdoiextendthis.jpg [View same] [iqdb] [saucenao] [google]
1737892

I need to extend these cables, never done any of it before. So lets say I'm at the store, what exactly do I pick up to get it extended?

>> No.1737896

>>1737892
Just get similar sized wires some heatshrink wrap and a soldering iron

>> No.1737899

>>1737892
For that gauge I would use scotchloks.

>> No.1737900
File: 1.77 MB, 3724x2096, D4E11457-C51B-4389-91A7-024C799A27D5.jpg [View same] [iqdb] [saucenao] [google]
1737900

Anybody here fix TVs?

Muh old Vizio. This one had intermittent problems where I would turn the power on, and the screen wouldn’t come on like no backlight power. But if I turned it off and on again it would work. It got worse over time, eventually I would have to turn the power on for like 45min (screen wouldn’t come on, but Vizio logo would light up) and after leaving it on for awhile, I could flip it off-on again and the screen would come on. Then it was finally killed.

Looks like there’s the power board in the middle, control center and shit to the right, and then everything for the picture at the top.

I don’t need to fix it, I already got a new one, but it would be fun to learn and troubleshoot it. Thought I would find a bulged cap or a bad relay because of that little click when the screen powers on, but don’t see anything.

>> No.1737902
File: 825 KB, 3724x2096, 3148D416-D1FE-4B17-A133-4C516AA2BF5C.jpg [View same] [iqdb] [saucenao] [google]
1737902

>>1737900
The two white wires running from the transformers to either side of the screen, they have the warning label, is that the backlight? And then picture goes through the ribbon at the top?

>> No.1737913

>>1737896
>>1737899
material of cables doesn't matter as long as its similar sized?

>> No.1737914

>>1737913
It's a fucking wire. You can use a glass full of salty water to connect them, similar size is because too thick is a waste and too thin won't handle the current

>> No.1737923
File: 38 KB, 1314x845, horn timer.png [View same] [iqdb] [saucenao] [google]
1737923

>>1737888
Nice numbers

So I came up with this 555 timer circuit to drive the horn for about 10 seconds.
The second 555 timer creates a pulse that should trigger a relais at the output so that it stays on for about 1 second and off for about half a second.
That tilt sensor is a good idea and simple enough. I just have to secure it so it doesnt trigger automatically.
I also added an indicator LED that stays on until I "unarm" the device to see if someone has been messing around with my plants.

Adding one of those arduino things to send me an SMS when it gets triggered would top it off but I dont think the signal will reach me.

Can I legally build a radio transmitter strong enough to receive the signal from my underground garage to my house, which is like 70 meters away?

>> No.1737931

>>1737902

yes and yes.
prob 1500V on the backlight wires, so hands off.

>> No.1737934

well ltspice updated and it broke. Anybody here use circuit labs? How does it compare?

>> No.1737939

>>1737923

the usual way to turn on the second 555 is to connect pin 3 of the first one to pin 4 of the second one. this eliminates the first transistor.

as for 70 ft, i have several wireless home phones that can do that. you'd wire a switch across the Locator button on the base, and get an alarm signal at the receiver. otherwise, steal a couple of FRS walkie-talkies from some kids.

>> No.1737943

>>1737939
Good idea.
Though how else would I trigger the indicator LED and have it stay on until I switch it off?

>> No.1737946
File: 1.70 MB, 3724x2096, E8F0D934-BC8F-4BCA-9114-1B0257123E12.jpg [View same] [iqdb] [saucenao] [google]
1737946

>>1737931
>1500V
Soooo don’t stick the Walmart multimeter with cheap ass leads into there?

>> No.1737960

>>1737946
>CAT III
Yeah no

>> No.1737967

>>1737960
Any leads to what would be fucked up? It died over time. At first, I could flip the TV on-off-on a couple times and the screen would come on. Then it took on-off ten times, then on for 30min with no screen, a quick off and back on and the screen would turn on then.

Otherwise I’m gonna rip the boards out for my scrap pile and trash that heavy ass thing.

>> No.1737996

>>1737943
>how else would I trigger the indicator LED and have it stay on until I switch it off?

you just feed the second transistor from pin 3 as well, thru 1K.
the 555 is able to source and sink 0.2A, so a buffer transistor isn't needed.

>> No.1737997

>>1737967
Got a service manual? Schematics? Don't bother trying to probe the HV side of that circuit. Probe elsewhere.
Are parts cheap? If so, you can try just dropping a new power board in it.

>> No.1738026

>>1737967
Caps. They don't have to bulge to be bad. Ideally throw em on an ESR meter. But just replacing all the electrolytics is likely not going to be that expensive.

>> No.1738030
File: 260 KB, 640x776, A18B376A-0380-43FF-93C0-43985880EBAD.jpg [View same] [iqdb] [saucenao] [google]
1738030

>>1737997
It’s a 10 year old Vizio 720p TV, I’m not spending money on it. I already got a new LG 1080 Smart TV to replace it for <$300.

>>1738026
I shoulda asked for that meter for xmas. What’s the go-to for an ESR meter that isn’t Fluke expensive but better than no-name Chinesium and I can actually trust the readings?

>> No.1738033

>>1738030
I thought the chinesium ESR meters were fine.

>> No.1738049
File: 308 KB, 640x995, F44A3A5F-E792-4029-8E31-FB1CC106F1D6.jpg [View same] [iqdb] [saucenao] [google]
1738049

>>1738033
Hmm, there is a $40 with shitty reviews, this looks like an updated version with better reviews. But that $120 meter looks like it has a ton more features that I don’t know how to use yet (L/C/R).

>> No.1738055

>>1737845
most people would like the current limiting to take effect before some time next year. go back to your containment general

>>1737997
shush, we're trying to get bepis to kill himself

>> No.1738058

>>1737923
>legally
probably not. the law concerns itself with things like like duty cycle restrictions and the purpose of the link as well as just power
but those little 433MHz transmitter-receiver modules are under a buck for the pair and you could send an intermittent tone over them. better read your local radio regulations. in burgerland you're looking for FCC Part 15

>>1737939
>FRS
he said legally

>> No.1738059
File: 330 KB, 1280x958, 7C6F3CF8-4335-418E-A20C-6FA657DEB7D4.jpg [View same] [iqdb] [saucenao] [google]
1738059

>>1738055
Hey I left you cunts alone for awhile, I was busy with work n the Latinas. Now I’m thinking of what I will do when I get my Playskool Blue n Yellow soldering iron.

I still have that Arduino kit from last Xmas, I was thinking about what I could do with it, maybe it could become a weather station?

>> No.1738060
File: 143 KB, 1000x1000, BM237R.jpg [View same] [iqdb] [saucenao] [google]
1738060

a while back i watched eevblog's multimeter showdown and it seemed that the brymen stuff was pretty good for the price. Thoughts?

>> No.1738062

>>1738060
I bought one of those china-only Fluke meters Ave talked about a few years ago. Pretty happy with mine.

>> No.1738065
File: 1.53 MB, 2293x3024, F35C848D-BE4C-43D0-B6B1-9523328CC3C3.jpg [View same] [iqdb] [saucenao] [google]
1738065

>>1738060
There are also the blue Brymen meters that were designed by EEVblog. They don’t look like cheap no-name trash at all. There’s a big difference between bottom tier Chinesium, mid-grade, and paying for the Fluke pedigree.

>> No.1738093

>>1738049
EEVBlog did a review of a Chinesium LCR ESR meter, seemed to be bretty gud

>> No.1738094

>>1738093
One of these? https://youtu.be/7Br3L1B80ow

>> No.1738096

>>1738094
Yeah, check his forums though, there are much more refined ones now apparently
You can build one yourself with an ardweenie too

>> No.1738113

ltspice is so fucking clunky, when will they update the UI

>> No.1738119

>>1738113
probably about the 11th of never
just use MicroCap

>> No.1738176

>>1738113
It's even worse on macos, you have to do everything with shortcuts because there is no toolbar button for wire or component or label

>> No.1738179

>>1738176
>macos
found the problem

>> No.1738180

>>1738179
xd

>> No.1738196
File: 32 KB, 600x324, Conducted_Electrical_Weapon_Taser_X26_360000.jpgE3E9C5D8-CB8D-4B54-8092-7BD2C87C53EBLarge-1.jpg [View same] [iqdb] [saucenao] [google]
1738196

Can someone please explain to me what kind of voltage these things actually produce?
There's so much misinformation like "it's not the volts that kill you, it's the amps" and other such bullshit that confused me about electricity for years and made me think voltage and current were two separately controllable properties rather than them being completely dependent on each other.
Obviously these things can't induce 50,000 volts across you or whatever people claim because due to ohms law they'd need to be capable of supplying a whole neighborhoods worth of current to make that potential difference possible.
So what is the actual sustained voltage and current that one of these creates across and passes through you?

>> No.1738199

Does anyone know how the sweep circuit works in old school tube oscilloscopes? I'm trying to repair one and I'm having trouble with the horizontal section which seems to be composed of a double triode and some passive components.

>> No.1738204

>>1738199
I imagine they're fairly similar to B&W CRT television circuits, though with the horizontal sweep time/slope being variable (dependant on time/div setting) and only happening on each trigger. Then the vertical deflection plates will be driven near directly by the signal.

Anyhow, just look for the service manual of your scope, should have a circuit diagram among other things. If it's using triodes anywhere instead of transistors it's likely quite the dinosaur, post pics please.
As usual with old pieces of equipment, capacitors and dirt are typical possible problems. Cracked solder joints could also be a potential issue, as could old overheated components.

>> No.1738206

>>1738196
>sustained
I don't know, but it is realistic to say they'll charge up to 50kV. They have a tiny capacitance that they bring to that voltage.

>> No.1738210

>>1738196
lel, I too am ameriflat-earther and enjoy pretending that the third dimension doesn't exist, let alone the fourth
it should look a bit more reasonable when you think in terms of coulombs, joules and microseconds. you don't need a full half-cycle of medium-voltage mains to depolarize some nerves temporarily

>> No.1738214

>>1738206
>>1738210
>It's a 50,000V device because there's 50,000V across some part inside it when it's not doing anything and when it's activated that voltage exists for a picosecond
Isn't this an absolutely retarded way to measure? It's like me claiming that I can lift a truck because if I held my arms up and you dropped one ontop of me there'd be a few milliseconds where I was holding it above my head.

>> No.1738217

>>1738214
It's marketing

>> No.1738227

>>1738217
Are there no regulations against this kind of thing?
Can I stick a massive capacitor into a 900W microwave that charges for 10 seconds and then blasts the food for 1 and then market it as a 9000W microwave?

>> No.1738231

>>1738227
I'm not a lawyer, but in my opinion it's not the same thing. The taser inherently works by charging to some high voltage and delivering that energy in a short time whereas a microwave is supposed to do sustained heating. But maybe you could sell a microwave with 9000W "pulse feature" if your marketing department is good.

>> No.1738232

>>1738214
>>1738217
I think the best way to do it would be to deliver as much power as you can, say, 2 times per second. You could pulse the current into the RF range, but then it doesn't cause a shock anymore, it just burns the skin.
I wonder if it would be legal to buy one of these and pack as much power as you could into the handheld form factor. That would be a fun project, you probably could char meat/skin with it.
Tasers are actually dangerous, cops do kill people with them. If you put one of these directly into the chest your heart would be stimulated just as your arm contracts if you shock your arm. But they just don't have the stopping power and don't hurt that much, they're pretty much just torture devices to get you to stop trying to defend yourself once you're already immobilized.

>> No.1738238

>>1738227
You could, but the R&D required to make a pulsed MW oven would cost more than the profits from the BS marketing. MW ovens already work at 2.5 kV, making them work at 25 kV without massive arcing would require much better insulation.

>> No.1738261

>>1738227
The main way truth-in-marketing happens isn't through regulation, but through criminal prosecution for outright fraudulent claims, class action lawsuits for very deceptive claims, unfair competition lawsuits for claims that disparage the competition, and "buyer beware" for mere puffery.

>> No.1738316

>>1738214
You're right, technically static electricity from your car door could be higher than that. I wonder what would be a better metric then, Joules or Coulombs (per jolt)? Perhaps a combination of one of those alongside time constant or power that's determined by human impedance? Because a 1kV capacitor with 100J stored in it will hurt more than a 10V capacitor with 100J stored in it, but I don't think that using coulombs alone is enough to take that into account.

As far as "what voltage do they use", I see chink taser modules and the like advertise with "1 million volts" and "2 million volts" and the like. Not that it means anything, of course.

>> No.1738317
File: 321 KB, 640x1096, 72677576-FC38-4B71-9E81-87292F1DEF8D.jpg [View same] [iqdb] [saucenao] [google]
1738317

>>1738094
I was confused how you’re supposed to test caps in-circuit with this, but I guess pic related explains it better.

>> No.1738326

>>1738316
>I wonder what would be a better metric then
you would need to measure the hurtz with a painometer

>> No.1738327

>>1738317
You can't really get a proper measurement in-circuit. You might get an estimation of capacitance and ESR but you can't test leakage which is what generally kills components.

>> No.1738347

>>1738316
means about as much as any other figure of merit on chinkbay

>>1738214
not quite. it'd be more like you claiming you can lift a truck by placing a long lever under it and free-falling at terminal velocity well on the other side of the fulcrum
voltage is meaningful beyond its effect on the erections of IDF-trained burger cops. it also roughly indicates how much air it can ionize in order to reach the target if direct contact can't be made

>> No.1738370

>>1737417
holding up li ion voltage is fine, current based trickle is bad

>> No.1738378

>>1737900
>>1737902
>>1737946
>>1738030
>>1738049
>>1738317
>4. /ohm/ is an anonymous, non-smoking general.
Thank you for your compliance.

>> No.1738456

>>1738378
>Thank you for your compliance.

not Nazi Germany, so compliance is optional.
stay free; fuck the haters.

>> No.1738490

>>1738055
>>1738055
What general? What the fuck are you talking about?

>> No.1738492
File: 172 KB, 1241x581, IMG_20191222_160822.jpg [View same] [iqdb] [saucenao] [google]
1738492

Can inductors leak? I thought it was just some coil embedded in resin or whatnot. This stuff appeared (at least I'm pretty sure it wasn't there when I changed the cooler) along the seams of almost all inductors on my GPU. It's glasslike and can be scratched off with a screwdriver.

>> No.1738494

>>1738492
It's glue, holding the two halves of the ferrite together. It's just sloppily applied.

>> No.1738499

>>1738494
That makes sense I guess, thanks

>> No.1738517
File: 5 KB, 400x400, 6 million hours in mspaint.png [View same] [iqdb] [saucenao] [google]
1738517

newfag looking at a pc psu, why does this capacitor not combust?
it says 0.47uF and it's across 230v AC
if i'm not mistaken P = U^2/ Xc = (230V)^2 *2*(pi)*50Hz*0.47uF = 7.8W
it's encased in plastic, ~ 2*2*1cm, how does it dissipate so much heat?

>> No.1738523

the power draw is probably related to internal resistance, not just capacitance

>> No.1738535

>>1738517
"If the load is purely reactive, then the voltage and current are 90 degrees out of phase. For two quarters of each cycle, the product of voltage and current is positive, but for the other two quarters, the product is negative, indicating that on average, exactly as much energy flows into the load as flows back out. There is no net energy flow over each half cycle. In this case, only reactive power flows: There is no net transfer of energy to the load; however, electrical power does flow along the wires and returns by flowing in reverse along the same wires. The current required for this reactive power flow dissipates energy in the line resistance, even if the ideal load device consumes no energy itself. Practical loads have resistance as well as inductance, or capacitance, so both active and reactive powers will flow to normal loads. "

>> No.1738539
File: 484 KB, 2048x1536, IMG_20191222_164130.jpg [View same] [iqdb] [saucenao] [google]
1738539

I finally finished the collar and did some field testing with my doggo, it went great, makes me think god exists and he's a cunt
>put collar on doggo and walk up a 500m long hill
>arrive at an edge of a dark forest
>grab doggo with intent to turn on the collar
>that fucker managed to lose it somewhere in 500 fucking steps
>walk up and down the hill for an hour looking for it with a flashlight
>nothing
>fuck it, I'll just make new one
>turn around to go back home
>take LITERALLY ONE STEP and feel a crack under my shoe
>no fucking way
>yes fucking way, i just stepped on the collar the moment i stopped looking for it and broke it
FUCK

>> No.1738590

>>1738535
I don't get the last half

>> No.1738594

>>1738539

your cunt god gave me a good laugh, so she's good for something.

>>1738590

just remember reactance is one of the Weirding Ways. no-one understands.

>> No.1738615

>>1738590
The energy that goes into charging the capacitor comes back out in the next part of the sine wave rather than being converted to heat and blowing up the component as in a resistor. V^2/Z is the RMS power, but remember that when you're looking at the RMS it's ignoring the sign. So half the time the actual power going into the component is positive, but the other half of the time it's negative.

>>1738594
brainlet

>> No.1738697

>>1738370
still putting current into the battery which is doing nothing good for you

>>1738456
shut up bepis

>>1738539
if you wanted to see the collar, you should have left it turned on

>> No.1738722
File: 85 KB, 877x744, layout.png [View same] [iqdb] [saucenao] [google]
1738722

I'm not really happy how the GND trace of my 5A switching relay goes through my VCC polygon.
Can that be a problem?
What changes would you suggest?

>> No.1738734

>>1738722
is it actually drawing 5A through any part of the copper that feeds sensitive circuitry, or does the current flow stop at the terminal block?
think of copper fills as a grid of several µΩ-tier resistors and decide how many scarce, valuable fucks you have to give about a few millivolts of noise. then think about how many fewer millivolts you would have to allocate your fucks across if you just mostly fill each layer with power and call it a day

>> No.1738758

>>1738734
You're right
I should call it a night

>> No.1738870
File: 1.56 MB, 4032x3024, E01E8E48-A734-419F-AADA-0904A199C515.jpg [View same] [iqdb] [saucenao] [google]
1738870

>>1738697
>bepis
Nope. See, it is still anonymous when you can’t tell which idiot is Bepis and which idiot is just another idiot.

>tfw I’m outta money to spend on gifts but still sort of want an extra gift for a couple people, so I might hack together some 2/3p 18650 power banks wrapped in duct tape
Merry Xmas! Enjoy you’re fire hazard!

>> No.1738974
File: 1007 KB, 1908x4032, 20191223_011714.jpg [View same] [iqdb] [saucenao] [google]
1738974

Retarded anon back at it again
I finished the circuit for my induction furnace but as I feared it doesn't work. The LED lights up but the fan doesn't turn on when hooked up. Hooking up a modified server power supply that has a maximum output of 900W just made smoke come from the PCB near the LED

I suspect my shitty soldering on the fuse boxes has something to do with it, but here's a pic of the burn
Any ideas? Ill post some pics of the unit and soldering work

>> No.1738975
File: 2.41 MB, 4032x1908, 20191223_012718.jpg [View same] [iqdb] [saucenao] [google]
1738975

Unit, work coil not attached and was getting power from the psu to the unit by alligator clips and a banana clip/XM60 adapter

>> No.1738978
File: 401 KB, 820x455, asdfwefaa.png [View same] [iqdb] [saucenao] [google]
1738978

Other than fixing the soldering job on the fuse holders I'm not sure what else could be wrong, here's a schematic.
Promise this'll be the last time i bother y'all about this

>> No.1738984

>>1738978
so, if I understand, the work coil's opposition to change in current flow is what turns each side of the circuit on and off to produce oscillations, and the slight normal variations in components help to get it started. without the work coil, the input will be shorted via each side's inductors and MOSFETs

>> No.1738997

>>1738974
You need a fast turn on PSU or the circuit will lock up and burn up. I think I got that from eevblog or some other youtuber.

>> No.1738999

>>1738517
Capacitors never dissipate any power at all. The power is dissipated at the source's internal resistance.

>> No.1739012

>>1738999
Ideal capacitors never dissipate any power at all. The power is dissipated in the source's, capacitor's, and wire's resistances, most of this will be in the wires (and probably transformers too) for a mains grid. An ideal source, capacitor, or wire has no resistance at all.

>> No.1739045
File: 389 KB, 2048x1536, IMG_20191223_142020.jpg [View same] [iqdb] [saucenao] [google]
1739045

I'm trying my hand at potting to waterproof this cunt
Wish me luck
It's some special premium expensive silicone without any acid so it should not fuck up the traces

>> No.1739110
File: 89 KB, 800x800, 5a17ff98-6304-4e94-a4c4-f4a26f2df3c4.jpg [View same] [iqdb] [saucenao] [google]
1739110

Hey guys im in need of a new multimeter, and want to qso which one you guys have and if you would recommend it?
>Pic semi related

>> No.1739111
File: 579 KB, 2178x2763, 121GW-KickstarterHeroShot1-crop[1].jpg [View same] [iqdb] [saucenao] [google]
1739111

>>1739110
This is the only acceptable product unless you want to be considered a joke by all electron connoisseurs

>> No.1739112
File: 103 KB, 1000x1000, Fluke-15B-Digital-Multimeter-Auto-Range-4000-Counts-AC-DC-Voltage-Current-Resistance-Meter-Capacitance-Frequency.jpg [View same] [iqdb] [saucenao] [google]
1739112

>>1739110
I got pic related on AliExpress last month for about $57 shipped. I like it. I don't have measurements of how accurate it actually is but I'm reasonably sure it's in spec.

>> No.1739144

>>1739110
>buyfagotry
>hasn't said anything about intended use
If you aren't using it for anything special then literally the cheap ones on Ali are ok.

>> No.1739167
File: 17 KB, 1084x622, Án titils.png [View same] [iqdb] [saucenao] [google]
1739167

why are massive boost converters so hard to come by? i've been looking to get every last horsepower out of my electric mountain board without stuffing it full of expensive batteries. the problem i'm running in to is that i can't find step up converters rated for more than 500 watts. would it be too hard to find the parts to build one rated for 6kw?

second. am i retarded for wanting 8hp from something weighing 20kg max? i'm running around 1,5hp (20v) on my prototype and it can be pretty scary as it is but i plan to eliminate rear steering among doing re-designs for stability

>> No.1739171
File: 1.74 MB, 3796x2520, progress.jpg [View same] [iqdb] [saucenao] [google]
1739171

>>1739167
i thought you guys might be amused to see my progress. planning to weld up a new one since i'm putting 8" scooter wheels on it

>> No.1739174

>>1739171
>hit a stone or bump
>die
based

>> No.1739180

>>1739167
>i've been looking to get every last horsepower out of my electric mountain board without stuffing it full of expensive batteries
Those "expensive batteries" are cheap compared to the route you're going where you'll have to chuck it in the bin and buy new ones every use due to over-discharging.

>> No.1739191

>>1739167
because boost converters are intrinsically not very efficient, and the energy wasted turns into heat. given a 92% efficient converter, where are you gonna put 500W of heat
the best way to overclock your scooter is to rewind the motor

>> No.1739223

>>1739180
i assume you're talking about the car batteries being over-discharged. yeah you're right i shouldn't be looking at a lead acid battery's ah even though i have more than 6 used ones at the moment i'd run out pretty fast after upgrading the board

>>1739191
500W doesn't sound like alot when placed on something that's moving at >35km/h. granted, i might have to place air scoops for greater cooling efficiency.

>> No.1739226

>>1739191
>>1739223
install a propeller in the front
extra propulsion+free forced air cooling

>> No.1739229

>>1739226
lmao this thing is dangerous enough as it is

>> No.1739237

>>1739167
Certain topologies are used only in certain power ranges

>> No.1739320

>>1739237
which topologies are used in the "mighty morphin" power ranges?

>> No.1739322

>>1739223
Waste heat is stored in the balls
Paging the carbon fiber ball warmer guy

>> No.1739325

>>1739320
why don't you just connect your batteries in series and stop fucking around

>> No.1739336

>>1739325
This, no reason you actually need any sort of stable voltage level from a switching converter when you could just stack 18650s to get a similar voltage. I don't know of any RC or other serious electric vehicles that use some sort of voltage converter between the battery and the motors, they all just run off the battery voltage.

>> No.1739337

>>1739320
If your drawing is correct and it's a dc/dc thing, the usual buck/boost converters are used. But usually they are custom built? I mean, what cosumer application uses 5KW and needs DC/DC conversion? The closer I can think off are suicide showers and here in brazil they are are either resistances with shunts or a triac controlled heater (and that is ac/ac) And this >>1739336

>> No.1739356

>>1739337
Makes you wonder if you could make a combined 3phase BLDC driver + boost converter, where each phase has what's basically a boost converter on it. Potentially useful if you have some sort of high-voltage (electrostatic or piezo?) motor to drive off a single cell.

>> No.1739372
File: 25 KB, 492x428, 1559358751844.jpg [View same] [iqdb] [saucenao] [google]
1739372

>>1739356
just rewind the motor

>> No.1739374
File: 16 KB, 710x410, Capture.png [View same] [iqdb] [saucenao] [google]
1739374

I'm doing an inverter as a part of a project, using IGBTs in a h-bridge configuration. Naturally when it comes to this kind of topology the switching does have some overshoot plus ringing tied to it, and it does show on the load albeit at reasonable levels (say half a volt switching noise on a couple dozen volts of output voltage). The issue that I'm having is that I'm using a voltage divider to monitor the output, and while it does attenuate the output waveform properly, for some reason the switching noise just passes through directly (in fact it seems to be carried to every goddamn part of the circuit after the h-bridge unchanged). Does anyone has any idea of what could be causing that? (and no, it is not the probes picking up EMI).
I understand that I could try adding some snubbers to try and mitigate/get rid of the ringing altogether but what I really want to know is how can the noise just go through like that.

>> No.1739440
File: 50 KB, 970x545, 1518327327406.jpg [View same] [iqdb] [saucenao] [google]
1739440

>spent over 100 bucks on PCB manufacturing and parts in a span of a day
is it really worth it?
the circuit not going to work anyway
I even posted this in the wrong thread out of stupor

>> No.1739443
File: 99 KB, 866x706, 1562374814828.jpg [View same] [iqdb] [saucenao] [google]
1739443

>>1739440
bodge happens. roll with it

>> No.1739444

>>1739374
What frequency is the switching? Also tr and follow the suggestions in this guide to eliminate the apparent ringing caused by the probe.
https://www.electronicspecifier.com/power/oscilloscope-probing-techniques-for-measuring-power-supply-ripple
My guess is that you have bad decoupling and/or power supply

>> No.1739449

>>1739444
probably capacitance where you don't quite expect it e.g. inside a resistor or inside the probe itself

>> No.1739451

>>1739325
yeah it's probably the best route to extract the battery cells and make a 48v bank. i'm tempted to wire them in series but i think i'll fry the battery controllers meaning i cant use these batteries in my tools and charging them wouldn't be as convenient

>>1739336
my biggest problem is the amperes. a 48v 120a battery pack would be massive and expensive so that's the reason i think i'll add supercaps

>> No.1739463

>>1739356
no

>> No.1739464
File: 4 KB, 600x287, symbol.png [View same] [iqdb] [saucenao] [google]
1739464

What does this symbol mean?
It's a resistor alright, but what about the two dots?

>> No.1739465

>>1739464
Looks pretty nonstandard.

If it's something in a textbook it might be to remind you that there's reactance in there?

>> No.1739466
File: 1 KB, 130x130, resistor-protection_2.png [View same] [iqdb] [saucenao] [google]
1739466

>>1739464
Protective resistor: It also functions as a fuse. Sometimes it's drawn like pic related.

>> No.1739467

>>1739465
Its on the silkscreen on a switchmode powersupply I'm troubleshooting.
The resistor in question is .1Ohm. Maybe acting as shunt.

>> No.1739468

>>1739466
That makes sense, thanks.

>> No.1739470

>>1739464
A jumper

>> No.1739472

>>1737652
Enlighten me, senpai.

>> No.1739474

>>1739472
Not him but a lot of companies give out free samples for students/teachers/randos if you
>explain what you are doing
>pay shipping
>don't ask for much (and still sometimes they can give out more than you think)
I almost got some Coil Craft high quality cores for a converter for a lab here in brazil, but the shipping here is so silly it was deemed easier to use local material and spin our own inductors

>> No.1739475

>>1739474
Just google
>company x + free samples
Usually they have a page with the info on how to ask. Some are cunts, other are quite nice (coilcraft was). There used to be something in the OP about free samples.

>> No.1739476

>>1739475
finna bouta google "rocketdyne free samples" lads

>> No.1739477
File: 55 KB, 932x524, Pigeons.jpg [View same] [iqdb] [saucenao] [google]
1739477

I'm making a small stepper controller board. Anything better than the L293?
I'm only interested in SMD. Thx

>> No.1739493

What method would you use to convert 50 Hz mains to 60 Hz mains or vice versa? Mechanically-coupled motor-generator set? Or something electronic? I've got some older audio shit I need adapting and I'm debating trying to build it.

>> No.1739494

>>1739444
>What frequency is the switching?
20KHz.
As I said, I'm pretty sure it's not bad probing (also the ringing comes from parasitic inductance/capacitance on the board itself and not the probes).
I also did think the supply could be the problem, but even when using a battery and a fairly large capacitor bank (> 1mF) the issue remains.

>>1739449
I thought the resistors might have been it and tried a voltage divider with multiple resistors to cancel out the parasitic capacitance (which isn't that much to begin with since I'm using SMD parts), but that still didn't do it.

>> No.1739503
File: 97 KB, 306x236, fusible.png [View same] [iqdb] [saucenao] [google]
1739503

>>1739466
>drawn like pic
4.7Ω 5% fusible
post baroque style

>> No.1739550

>>1739374
Post measurement set up

>> No.1739566

>>1739550
honestly this, you're not gonna silence people who think your probing is bad unless you provide proofs

>> No.1739594
File: 15 KB, 456x320, ccccccccccccccccccccccccccccccccccccccccccccc.png [View same] [iqdb] [saucenao] [google]
1739594

on a lm386, if I'm not using it to drive a speaker (as in, using it just as a preamplifier), do I need the giant bypass cap on the output?

>> No.1739598

>>1739550
>>1739566
Never mind then, I'm currently on vacation so I don't have access to my setup atm. Have a nice one you guys.

>> No.1739600 [DELETED] 

>>1739493
>convert 50 Hz mains to 60 Hz mains or vice versa?

for audio? that's crazy. 99.99% chance your gear will not notice any difference.

>>1739594

if you power it from a 9V battery, there will be 4.5Vdc on pin 5. this will draw massive current thru the speaker and cause mass distortion unless you block the DC with a fat-bottomed cap.

>> No.1739602

>>1739493
>convert 50 Hz mains to 60 Hz mains or vice versa?

for audio? that's crazy. 99.99% chance your gear will not notice any difference unless it's something like a turntable, using an induction motor. if so, you could use an inverter that generates 50 or 60Hz from 12Vdc.

>>1739594

if you power it from a 9V battery, there will be 4.5Vdc on pin 5. this will draw massive current thru the speaker and cause mass distortion unless you block the DC with a fat-bottomed cap.

>> No.1739603

>>1739602
I'm not using it to power a speaker, I'm using it to amplify a signal being sent down to an lm3916 LED VU bar graph driver

>> No.1739608
File: 61 KB, 560x413, retro_vu_meter_schematic.jpg [View same] [iqdb] [saucenao] [google]
1739608

>>1739603

then you need to some some more processing: first, use the cap to remove the DC bias, then use an op-amp to rectify the signal, then a cap to smooth it. plenty of example circuits exist, so no need to re-invent wheel.

>> No.1739630

>>1739598
It's troubleshooting 101
>thing is made out of A, B and C
>Guy says A is ok
>says B is ok
>C is also ok
>still malfunction
Clearly something was not ok. Prime example was LED+buck anon who was so ashamed he missed something simple that he offed himself.

>> No.1739635

>>1737362

I'm not a regular. Should I ask about cable connections here or in /qtddtot/?

>> No.1739654

>>1739635
Porque no los dos?

>> No.1739664

>>1739654

Missa no speak lucha.

>> No.1739671

happy xmass you pitiful cunts

i hope you wrote santa for a new 4 channel scope an a fluke

>> No.1739704

>>1737652
>it's okay to be a merchant as long as you're the merchant
gtfo

>>1739477
DRV8844 in HTSSOP-28?

>>1739594
>using an LM386 as a preamp
seriously, y tho

>>1739671
I already know I'm getting nothing but a lump of carbon composition resistors :(

>> No.1739705

Which one is better a t20 clone or a Hakko 888d?

>> No.1739706
File: 338 KB, 600x440, 1569029709438.png [View same] [iqdb] [saucenao] [google]
1739706

>>1739664
se habla reddit, cabron

>> No.1739714

>>1739705
T-20 has a modest variety of tip shapes. T-18 puts the heater too far from the tip to push a heavy ground plane. it's a tough call that depends on the nature of your work, and you gotta make the call

>> No.1739743

>>1739706

Άντε χέσου ρε μαλάκα. Κάθομαι και σπαταλάω την ώρα μου με τον χαζό του χωριού.

>> No.1739749

>>1739704
I mean, what else would I use as a basic amplifier? My phone's output is too low to drive the lm3916 so I just set up an lm386 because they're audio amps that I have on hand

>> No.1739753

well lads i just saw the new star wars movie and let me tell you, it has nothing to do with electronics generals

>> No.1739776

>>1739753
Was there any funky physics in there, like the previous film's dropping bombs onto a star destroyer in space?

>> No.1739779

>>1739776
>dropping a bomb
>a guided torpedo with an engine on it

>> No.1739780

>>1739776
Sorry, I was also assuming you meant a new hope, not one of the new movies. Maybe they did do something shitty in the new movies

>> No.1739795
File: 122 KB, 609x878, 1573994503571.png [View same] [iqdb] [saucenao] [google]
1739795

>>1739749
think inside the box, Pic related
>basic amplifier
literally any non-power opamp that runs on the supplies you have available

>> No.1739818
File: 72 KB, 1146x713, 1574599713936.png [View same] [iqdb] [saucenao] [google]
1739818

>>1739749
>>1739795
okay maybe not a 358 type

>> No.1739836

>>1739440
sure it's worth it.
I spent $150 developing my last project. as far as hobby expenses go, that's not bad.
If you're broke, then, well, you need to choose cheaper projects. But if you've got a job, 100-200 bucks every now and again isn't unreasonable.

>> No.1739853

>>1739440
>>1739836
What are you making? I'm doing a bretty gud headphone amp and I'm not expecting to spend more than $60

>> No.1739869

>>1739853
not him but sometimes you go through a lot of changes in a project
>wanted to fold some serial-bus periphs into the micro, needed more/better pins/memory/periphs
>destroyed three prototypes when I got bit by https://www.analog.com/media/en/technical-documentation/application-notes/an88f.pdf
>wanted to add a version with integrated power LED driver
>wanted to isolate the LED driver and the smart chip, couldn't use shitty optos to do it without adding another micro on the LED driver side
>thought I would need inverter for opto on driver-side micro
>couldn't fit floating point commands for my DSL into the micro
and at some point you just stop and live with what you got until such time as someone pays you several grand to automate their state-licensed dudelmao greenhouse
>can't map both I2C ports on the new micro, will need to bit-bang
>would like to delet USB port for space
>didn't actually need inverter for optos on driver-side micro because the tiny will do it for me, now have to bodge over inverter footprint or insert buffer gate
>need to bodge pulldown resistors to reduce ripple on dimmer
>would like to add 5V out on the driver board for ease of ST-link programming, but oh well
>maybe it would be fun to move the STM32 to the LED driver side and put the isolation on the field bus side
silver lining: now I'm literally swimming in STM32 dev boards that accept 6V-48V power in, and can bodge a totally generic power LED driver for use in lamps etc. out of the older rev of the LED driver board

>> No.1739877
File: 1.92 MB, 4032x3024, 9C9E2139-D012-4E81-A989-E9EE913D7514.jpg [View same] [iqdb] [saucenao] [google]
1739877

>>1739671
>tfw I woulda bought myself the scope the past couple tax returns but the girl is too expensive
Not that she is so expensive, but moreso having the girl around forces me to spend money like an adult.

>> No.1739879

>spend like $80 on a project
>never build it because the fundamental idea is flawed and I got a superior replacement for free
>now have 10 rotary switches lying about and a big wasted piece of copperclad I was going to use as a faceplate

>> No.1739937 [DELETED] 

>>1739743
یک کیره گنده و کلفت تو کونت فرو میکنم تا تو باشی حرف نامرتبط و بیهوده توی ین میدان بافتن سبد زیر آبی نزنی.

>> No.1739987

>>1739879
the empire grows.

>> No.1740008

>>1739795
>>1739818
call me a retard but I still don't get it

>> No.1740017

>>1739879
>I have junk, what do?
Get rid of it.

>> No.1740021

>>1739795
You really only need the resistor to V+ as the resistors inside the chip form the bottom half of a voltage divider.

>> No.1740133

>>1740008
my point, tard, was to supply your own reference voltage to the 3916, somewhat below the default minimum 1.2V

>>1740021
diffused resistors tend to have a wide tolerance in absolute terms. solved with a trimpot

>>1739879
the copper clad can be chopped up for smaller circuits or front panels
the rotary switches are slightly less reusable

>> No.1740222

>>1739603
You're supposed to use a normal op-amp for this, not a dedicated power/speaker amplifier.

>> No.1740246

>>1739869
>can't map both I2C ports on the new micro
I had a work project quite a few years ago with an STM32F103 where using FSMC for an LCD display also remapped the I2C1 pins. I never did figure out if there was some way to re-re-map them, I just had the hardware guy jumper both I2C ports together to the same net because it was the easiest thing to do.

>> No.1740256

So what is the best STM32 variant to use?

>> No.1740265

>>1739853
I just finished the BOM for my headphone amp and it came to $60.21. $20 of that is ICs for my power. A $7 switching regulator to get +/-15V from +5, then 4 $3.5 linear regulators to get +/-12 and +/-5 off that. The 12s and 5s are for ICs in my analog signal path. I went with expensive LT parts because I want to reduce the ripple on the power for my analog ICs, but I'm thinking any switcher that's running up above 20kHz is going to be inaudible as long as I leave plenty of headroom on my analog parts. Anyone want to weigh in and tell me if I'm missing something?

>> No.1740272

>>1740256
What is your application?

>> No.1740273

>>1740272
Does saying "general purpose" make you angry at me? Is there anything much I'd be missing out on by going for the cheapest one?

>> No.1740274

>>1740265
That sounds like the ideal PSU setup for any audio circuit, could you tell me what split-rail switcher you're using? Can it run on higher or lower voltages?

>> No.1740275

>>1740265
Why use a switching power supply in the first place? LDOs are much much less noisy. Consumer audio gear always uses them for this reason. They waste more power but it's proportional to the current draw of your headphone amp which should be pretty low.

Is that BOM from a website like Digikey? Parts would be a quarter of that on Aliexpress but there's a 1000 piece minimum order on almost everything.

>> No.1740277

>>1740273
It all depends on what you need and want to do. You could buy a stack of blue pill boards from China, or you could buy one of various Nucleo boards.

>> No.1740279

>>1740256
the one that most closely matches the set of qualities your particular application requires
>general purpose
or, if you're thinking prospectively, whatever's on sale, lel
last time I did an LCSC order, I bought twenty STM32F302CB because they were actually cheaper than STM32F103CB at the time and the extra RAM sealed the deal. when I saw there are comparators and opamps too in there, I kinda wish I'd designed my current project to make use of them, oh well

>>1740265
>$7
probably about right
>4 $3.5 linear regulators
look up PSRR of the analog ICs and decide how much you need to care about electric vs. magnetic noise. if the former is small enough, you might just put the supply in a can, and outside of the can some series LC might be enough to keep the rail clean
>inaudible
there is the small matter of aliasing of the power supply frequency with input, but probably not a big enough deal to matter
>ordering a BOM from Digikey
wew lad. maybe you should think chink and redesign for aZn parts

>>1740246
I don't need six gorillion units (yet) and I'm not feeling especially motivated to spin another rev of this board so I think that's how I'm gonna bodge this. software I2C has many undesirable qualities

>>1740275
looks like he's got some power supply constraints, namely, can only depend on having +5V available. I'd be unsurprised if he also had a USB-to-aux converter somewhere in the box

>> No.1740292

>>1740274
ADP5071. Input voltage range is 2.85 to 15V.

>>1740275
>>1740279 is correct, I have 5V off a USB connector

>>1740279
>aliasing of the power supply frequency with input
To me, this is the heart of the question. What do you mean by aliasing? I've always thought of it as a problem strictly in the digital realm, or when moving from analog to digital or vice versa. I can't imagine any mechanism that a ripple at 100kHz could introduce a frequency component below 20kHz in, say, an op-amp circuit.


Building a DAC to pair this with is a project for the distant, distant future. The main issue is getting a 24 bit audio stream data from computer/phone software to USB to a codec to a DAC. It's a lot of steps that I don't have experience with.

>> No.1740298
File: 7 KB, 474x266, 1570414292434.jpg [View same] [iqdb] [saucenao] [google]
1740298

>>1740292
maybe mixing is the more correct term. any time you put two frequencies through a multiplier (and amplifier offset voltages and open-loop gain do vary according to input voltage because transistors are non-linear), sum and difference frequencies are going to happen
if your power frequency level is over 100000 then I'd only be a little concerned over the very remote possibility of fS noise coming from the DAC and mixing with fSW of your switcher, then mixing again with the difference frequency, generating a nice screechy 4kHz whistle under your 48kHz playback. it would take a perfect storm of shit for that to happen so don't sweat it, just use LC and RC filters to keep your rails clean (maybe leave some spare pads and insert some 0Ω jumpers) and eat a healthy diet rich in vegetables. in case it does happen it shouldn't be too hard to patch out

>> No.1740299

I need an incredibly stable 3.3v supply from an iffy 5v supply. The voltage regulators on the teensy Im using are alright, but if I wanted something more stable how would I achieve that.

>> No.1740307

>>1740299
define stable
and define incredibly
while you're at it, define your load

>> No.1740309

>>1740299
>>1740307
desu, I'm also curious about the 5v supply and why it's "iffy". If a linear regulator worked relatively fine for you, I can't imagine the output requirements being too strict or the input being that bad.

>> No.1740325

>>1740292
>2.85 to 15V
OH YES

>> No.1740386

>>1740299
Most likely use a better linear regulator than the ones on the teensy, if your 5V is stable enough that you're sure you'll be above the required voltage. If not, boost to a higher voltage then linear regulate. The more voltage the regulator is dropping, the better the PSRR.

>>1740325
What do you need it for?

>>1740298
Cool, thanks

>> No.1740401

>>1740222
Understood. Is there a specific reason why I should be using just an opamp instead of the full on amplifier?

>> No.1740412

>>1740401
Relevant criteria for amplifier ICs are
>power
>noise
>cost

You can't max out all three. When you use a chip designed for power applications, the other two criteria are taking a hit.

>> No.1740418

If you turn a pot in a buck converter to change voltage, how quickly does the buck react on average? as in how fast will the voltage raise or fall

>> No.1740422

>>1740418
depends a lot on the output capacitance and control voltage filtering. you can assume a hard upper bound of 100ms if that helps.

>> No.1740425

>>1740299
the term you're looking for is "line regulation" and "power supply rejection ratio", as in, how well the output voltage stays stable as the input voltage fluctuates. check for this in your regulator's datasheet.
Use all the capacitors recommended by the datasheet's application circuit.
Make sure your 5v supply doesn't dip below the regulator's dropout voltage.
If you do all that you will have a very stable 3.3v.
better question is why do you need better stability than the onboard regulator already gives you? Are you trying to use that as an analog reference voltage?

>> No.1740462

>>1740401

the fact the the LM386 is
- in your possession
- single supply
- idiot proof
- familiar
makes it ideal, in fact perfectly ideal, for your application. all other objections are trifles by comparison.

>> No.1740487

>>1740462
I can't tell if you're being facetious. I'm still trying to learn all of the caveats of implementing these circuits and a lot of these things don't yet come naturally to me.

>> No.1740493

>>1739877
Literally just got one of those last week. Best damn station I've ever used.

>> No.1740548

>>1740487
Well he's right, but only if you want to make something in a hurry. Common operational amplifiers have much higher open-loop gains than a 386, making them ideal for filters, preamps, and higher frequency purposes. If you try to pull too much current from one you'll end up degrading its performance and likely introducing some distortion. This is what power amplifiers were designed for, though these days its more common to use a class-D. Op-amps are meant to be very multi-purpose, so I'd much rather have a bag of 20 op-amps (which I do have) over a bag of 20 power amps. Also you get two in one DIP8, four in one DIP14. The split supply problem can be worked around in a few different ways, simplest of which is a voltage divider and a couple big caps.

So if you're buying any parts any time soon, get a handful of op-amps. Comparators too, if you plan on using positive/no feedback. Grab some radial inductors too if you don't have any, they're fun.

>> No.1740572

>>1737362
If my multimeter only has two ports can I still measure the resistance between the two to test the fuses?

>> No.1740574
File: 125 KB, 600x600, 13834-01.jpg [View same] [iqdb] [saucenao] [google]
1740574

Does anyone know of a small breadboard-compatible switch with a similar accusation force / feel to a mech keyboard switch?
I really hate how hard to press / noisy standard tactile buttons are.

>> No.1740576

>>1740572
It reads everything as 3 amps

>> No.1740583

>>1740574
there are silicone tact switches but they don't have the feel you're looking for
you can chop the bosses off of the bottom of a mechanical keyboard switch but you might have trouble getting them on-grid
actuation force is a parameter you can choose, just gotta shop a bit. the really crappy ali cheap tact switches are 260g, maybe you want closer to 180 or 150 or less

>>1740576
pop it open and see what you find. it's not gonna arc flash you if it's not connected to several hundred volts

>> No.1740591

>>1740576
>It reads everything as 3 amps

i had a meter from radio Shack that would show 3 for ''open circuit'' and for overflow conditions.
the 3 looks a bit like infinity, so that's what they used for infinity.

>> No.1740593

>>1740583
I already did, wasn't anything obviously wrong

>> No.1740694
File: 2.24 MB, 1920x1080, IMG_20191226_214901.png [View same] [iqdb] [saucenao] [google]
1740694

Alright, I'm re-implementing my lm3916 amplifier using an lm358 op amp.
When I was testing it with regular audio, I was getting crazy transients several orders of magnitude larger than the amplified audio signal. Now I'm testing using a ~130 Hz sine wave, with a gain of 2, and there's anomalies at what would be the negative maximum of the wave. What am I fucking up here? I think it might be due to the audio input's negative portion exceeding the max input negative voltage of -.3 volts.

Do I need to DC bias the audio signal input, then remove the DC offset after amplification? Or should I just rectify it before the signal even hits the op amp?

>> No.1740696

How can I get good at understanding wiring diagrams?

>> No.1740697

>>1740694
show your schematic

>> No.1740699

>>1740697
give me a minute

>> No.1740701

>>1740694
Looks like your signal is DC biased incorrectly, and at the negative peaks it's surpassing a transistor's bc reverse voltage and causing an anomaly at the collector. Chuck a cap in series with your input wave and tie the output of the cap to both the V+ and V- rails with resistors.

Nice scope btw

>> No.1740703
File: 13 KB, 413x374, 1574656660197.png [View same] [iqdb] [saucenao] [google]
1740703

>>1740694
and yes, you should dc offset the input to the op amp with something like Pic related, and do the same on the output between the amp and 3916 if needed

>> No.1740704

>>1740701
>>1740703
R and C values should be selected such that the resulting high-pass filter has a corner frequency below ~20Hz.

>> No.1740708
File: 34 KB, 1465x543, cccccccccccccccccccccccccccccccccccccccccccccc.png [View same] [iqdb] [saucenao] [google]
1740708

>>1740697

>> No.1740711

>>1740548
>simplest of which is a voltage divide

It's common to use an op-amp based rail-splitter too! adding credence to the "i prefer a bag of op-amps" idea.
of course, If I was buying shit on line anyway, I'd throw in some TI tle2426 and be done with it.

>> No.1740714

>>1739879
What kind of project needs 10 fucking rotary switches?

>> No.1740716

>>1740714
A bunch of cascaded 74HC164 ring counters where whether Q0-Q7 goes into the serial input can be selected with the rotary switch. Hooked up to a 4MHz or 32.768kHz crystal, or an external source.

>> No.1740746

>>1740708
What are the supplies for the op-amp?

>> No.1740747

>>1740746
also 15 volts, but the lm358 is single supply

>> No.1740755

>>1740694
Tek 2445?

>> No.1740769

>>1740716

Ahh. Interesting.

I actually need of a bunch of those switches since I'm making a few timers for things I need to control from about 30 seconds to 3 hours or so. One to control my coffee grinder time, another to control a fan that I forget to turn off, etc. I just want to push a button and it counts down the time to what it set on the switches.

They're incredibly handy to have for a bunch of things.

I'm using a few old tms3450nl chips to get the seconds out into a counter.

>> No.1740770

>>1740769
nah these days you just throw a rotary encoder in since micros are so cheap

>> No.1740771

>>1740769
>One to control my coffee grinder time,
the pros use a 6-pin micro and a pot

>> No.1740772

>>1740771
>pot
shiggy

>> No.1740773

Anyone do any wireless battery powered sensors that are oshw? Some kind of industrial monitoring with sensors that can run for 5 years.

I see moteino, ruuvi, waspmote, and anarduino.

Ruuvi looks to use too much power in sleep mode at 1.2uA and uses bluetooth... Anarduino is schematics only (no layout) per https://lowpowerlab.com/forum/moteino/moteino-vs-anarduino-mini-vs-canique/..

I found waspmote, but while they call themselves open source here:
>https://www.cooking-hacks.com/documentation/tutorials/waspmote#waspmote_vs_arduino

They don't provide schematics for their boards as referenced by a few posts in the forums.
>https://www.libelium.com/forum/viewtopic.php?f=20&t=26257&p=82453

So in this case open source just means you get cucked (software only).

Moteino can do 400nA. More of a hardware platform than a solution, but the community is active. Looks like that's the way to go.

>> No.1740789
File: 29 KB, 400x351, s-l400.jpg [View same] [iqdb] [saucenao] [google]
1740789

I've been meaning to learn electronics because it's the one trade I don't understand at all. Found pic related at the thrift store and I'm inspired to finally get started learning this wizardry.

>> No.1740791

>oh boy, i have a vacation this whole week, 9 days in total, i will get so many of my projects finally finished and progress a ton
>I'm on the 6th day today
>all i did is solder like 3 components and connected a few wires, so about 30 minutes of work in total
What the fuck? I legit can't figure out where the fuck did the rest of those 5 days went

>> No.1740792

>>1740789
Throw that into the trash. go on ali and order an arduino with a sensor kit and learn how to wire them up and make them work, it's easy, fun and you can make actually useful things with it.
If you just make a bunch of voltage divers to learn how resistors work for no other purpose than learning it will get boring fast and you will lose interest

>> No.1740793
File: 265 KB, 911x1205, 1574953453947.jpg [View same] [iqdb] [saucenao] [google]
1740793

>>1740772
y not tho

>>1740773
1µA is a pretty good standby current. anything more than that takes work and attention
>he thinks not getting layouts on demand is cucked
we don't serve lazy-ass bums here. go to the duino thread

>>1740792
>muh duino
cuck

>> No.1740795

>>1740792
Yeah maybe if he wants to learn how to be a fucking code monkey. Resistors, capacitors, diodes, transistors, op-amps, that's where the fun and innovative circuits are.

>> No.1740796

>>1740791
leisure is valuable too
but yeah I got a lot of solder I need to melt tomorrow

>>1740795
otoh if he just wants to sparky (notice the word "trade") none of this is going to help him

>> No.1740797

>>1740796
>leisure is valuable too
not if a project that should take you 3 days to do takes you 6 months because you procrastinate so much you basically don't do anything ever

>> No.1740798

>>1740791
>been a neet for the past month or two after uni finished
>haven't soldered for three weeks
nothing like having hobbies that you never get around to working on to induce self-hatred

>> No.1740801

should i led my soldering iron take naps bewteen 15 min soldering jobs, or should i keep it sleep deprived and on constantly when i don't use it for say 20 minutes?

>> No.1740803

>>1740798
I really hate my brain, i WANT to work on those projects but it just won't let me, it's always like "breh just watch one more movie and then you get the work done" rinse and repeat

>> No.1740804

>>1740797
that does suck. leisure is too valuable to piss away on not getting shit done :^)

>>1740801
according to Daoism, a man only has a certain number of orgasms before he dies. therefore, at least turn it down to 180°C

>>1740803
>just watch one more movie
saw Tropic thunder for the first time tonight. never go full retard

>> No.1740824

>>1740801

a 15-20W iron can be left on 24/7 but as you go up to 35+W, they start to create crud on the tip which becomes increasingly hard to remove. so, you need to turn those off, or keep cleaning them every 10 minutes.

this applies to quality tips in quality irons. if you have shitty copper or brass tips, any run time is bad for the tip, and bad for the limited lifetime of a cheap iron.

>> No.1740837

>>1740801
depends on heat.
I keep mine set to 320 or so, on the low side, and I can let it sit for long periods without it getting oxides on the tip.
If you run hotter, or have an iron that isn't temperature controlled, it might get crud buildup.

>> No.1740846

>>1740792
I already know coding and have made some Arduino stuff before. I want to learn the fundamentals of electrical circuits.

>> No.1740857

Is it safe to have my arduino connected via usb cable to my pc when i have things like i2C chips connected to it which are powered by a separate power source?
since both my computer PSU and the power source for the i2c chip which is a different PSU are non isolated power sources, will that fuck my shit up?

>> No.1740865
File: 578 KB, 2048x1536, IMG_20191227_162814.jpg [View same] [iqdb] [saucenao] [google]
1740865

It worked, neat!
I can now control the buck converter output voltage from an arduino with the epot.
Now I need one more epot to hack into the current control as well and the heart of my bench power supply will be complete

>> No.1740904
File: 1.69 MB, 2759x2604, IMG_20191227_135720.jpg [View same] [iqdb] [saucenao] [google]
1740904

Ok so i bought

>Platt, Make: Electronics

And i have a question in one of the circuits, pic related

He says that once one of those 4 NC contacts is open, the tension on the base of the transistor will go up, and the relay coil will be activated, so the realy stays turned on, but you need to add a diode (1N4001) to prevent reverse tension in the emitter of the transistor

But how is it possible that such reverse tension could exist? Isn't the emitter of the transistor silicon doped, and has electrict negative charges? Its a NPN by the way

Surely the resistence to go trough the emitter is higher than the coil of the relay? What am i missing

>> No.1740911

>>1740904
Sopa de macaco uma delicia.

>> No.1740912

>>1740904
This circuit doesn't make much sense to me. When there's no current in the relay coil, the armatures are in the positions shown in the picture, right? That position is going to drive current through the relay coil so the armatures will move and cut off current until they return back to their pictured position. So the voltage across the alarm is going to be oscillating probably a few hundred times a second. I hope that's not intentional because that's a really stupid way of making an oscillator.

Related to that oscillation, the voltage at the cathode of the diode will spike when the relay opens, as the coil needs to draw the same amount of current in that instant. The base-emitter junction is also a diode oriented in the same way, but maybe the diode is necessary to help block that voltage.

The doping of the transistor isn't really relevant here. While the emitter does have negative mobile charges, it also has positive fixed charges in the nuclei of the dopants, so about 0 net charge. You don't need to consider how transistors are doped unless you're designing semiconductors, you only need to consider the function of an NPN vs PNP.

>> No.1740914
File: 216 KB, 1000x943, 1577466439535.jpg [View same] [iqdb] [saucenao] [google]
1740914

>>1740904
Shouldnt that diode be positioned like so?
But desu I dont understand the point of either diode.

>>1740912
>That position is going to drive current through the relay coil
Not him but why?
The buttons in this position shunt the base current to ground leaving the transistor closed.
And why the oscillation you mentioned?

>> No.1740915

>>1740912
>the armatures are in the positions shown in the picture, right?

No, the contact that is feeding the coil goes to that position once the coil is energized, its meant to keep the relay feeding the alarm once its trigged, there is no ocillation

I do not get it why there would be this reverse tension on the emitter of the transistor, and why the need for the diode

>>1740911
sir please

>> No.1740919

>>1740914

>Shouldnt that diode be positioned like so?

Yes, he does add this diode on the relay coil later, to prevent flyback voltage, which i did understand after reading about it

The one in the original pic i still dont get too

>> No.1740923

>>1740914
With the armature as seen in the picture, the cathode of the diode is shorted to 9V. According to the OP the armature is there when the coil is energized so it won't oscillate at least.

>>1740915
So the only way to turn off the alarm is to remove your 9V supply? If you open circuited your 9V you'd have a voltage spike on the cathode again which you might want to protect the transistor from.

>> No.1740924

>>1740915
>sir please
I don't see much sense in that diode. I don't get the point of the circuit because honestly it's eye cancer. But when switching a relay you need a diode across it to avoid overvoltage, it's easy to understand when you look how inductors and capacitors behave. (Their differential equations) .

>> No.1740927

>>1740755
2465 actually

>> No.1740930

>>1740923
So its because the coil of the relay will act as an inductor once the supply is removed? Hm ok i guess i get it

This feels like a rabbit hole, every small question i have i need to go read several pages of theory to fully understand this, damn

>> No.1740932

>>1740930
No. The relay coil IS a inductor. It acts as a voltage source because it stores energy in the magnetic field.
The inductor equation is:
V=di/dt * L
>if the current changes very rapidly, the voltage rises (in magnitude)
>if the relay is active and you shut it down, you change it's current very fast, the diode is there because it conducts after it goes above 0.7v and makes it stay there

>> No.1740933

>>1740932
Play around in falstad and see this happen.
https://www.falstad.com/circuit/circuitjs.html

>> No.1740935

>>1740932
>>1740933

Very nice, easy to visualize it now

Thank you

>> No.1740971

>>1740904
The diode is because the relay coil will give a reverse kickback voltage when turned off (just like in a switching power supply) which can damage or destroy the transistor, or even rose a microcontroller output pin. But it's usually done directly across the relay coil as in >>1740914

>> No.1740972

>>1740971
rose -> worse

>> No.1740974

https://www.falstad.com/circuit/circuitjs.html?cct=$+1+0.000005+10.20027730826997+50+5+50%0A178+240+176+384+176+0+1+0.2+0.028647343075203644+0.05+1000000+0.02+20%0Ag+240+288+240+304+0%0AR+240+128+240+32+0+0+40+5+0+0+0.5%0Ar+384+192+464+192+0+100%0Ar+240+224+240+288+0+10%0Aw+528+192+528+288+0%0Ag+528+288+528+304+0%0At+144+192+176+192+0+1+-2.8363706004953997+0.6820849140402132+100%0Ar+144+192+144+128+0+10000%0Aw+240+176+176+176+0%0Aw+144+192+144+224+0%0As+144+224+144+288+0+1+false%0Aw+240+128+240+176+0%0Ad+176+208+240+208+2+default%0Aw+144+128+240+128+0%0Ar+144+288+240+288+0+1000%0Al+464+192+528+192+0+0.001+0.04997501249375311%0Ao+2+64+7+4099+5+0.1+0+1+0.625%0A

>> No.1741023 [DELETED] 

alright I'm beginning to think that negative over-drawing the noninverting lm386 input pin from the audio signal might have borked my lm386
I don't want to fuck up another one even though they're worth pennies

>> No.1741025 [DELETED] 

alright I'm beginning to think that negative over-drawing the noninverting lm386 input pin from the audio signal might have borked my lm358
I don't want to fuck up another one even though they're worth pennies

>> No.1741027 [DELETED] 

alright I'm beginning to think that negative over-drawing the noninverting lm358 input pin from the audio signal might have borked my lm386
I don't want to fuck up another one even though they're worth pennies

>> No.1741028

alright I'm beginning to think that negative over-drawing the noninverting lm358 input pin from the audio signal might have borked my lm358
I don't want to fuck up another one even though they're worth pennies

christ all fucking mighty I've made this post 4 times, I'm trying to say I fucked up the lm358

>> No.1741032

>>1740803
>>1740804
For me it's "just watch one more youtube video on project related, then get started on said project"

>> No.1741036

>>1740789

Do not listen to
>>1740791
For he wears a fedora and has an ironic feminine beard.

I learned all my early wizardry from Radio Shack 200-in-one learning kits and the Forest M Mims books. Great stuff. Follow your path anon, learn things, don't let the Tarduinos impede your adventure.

>> No.1741044

>>1741036
Thanks. I want to just make my own basic and helpful tools around the house and diagnose broken electronics. I have a lot of ideas and I know they are doable, I just never learned electronics as a kid. I have about everything else in fabrication under my belt though.

>> No.1741067

>>1741028
You need to post your schematic

>> No.1741068

>>1740798
I only work on my hobbies when I have university stuff to do. It's my version of procrastination

>> No.1741077
File: 1 KB, 334x191, dc bias.png [View same] [iqdb] [saucenao] [google]
1741077

I've been looking at how to bias my input signal. I see the design in pic related, but since it's not a traditional voltage divider I can't just use the voltage divider equation. Is there an equation on this one to find the values of the resistors?

>>1741067
I did, it's >>1740708

granted now I'm trying to work on biasing the input so the schematic's going to change a bit.

>> No.1741080

>>1741077
nevermind I'm retarded, I just voltage-divide the 15 volts first with a standard divider and then use that as the biasing voltage.

>> No.1741084

>>1740904
>putting loads on the emitter
that cancerous book is going to be removed from the OP next time I get my hands on it

>>1741028
yes, that is quite possible
if you need the single supply, add a large cap in series with the input, about 10µF, and add a voltage divider, like >>1740703 pic related. connect that to a non-inverting voltage follower (+ on input, - on output). feed the output from the voltage follower into the rest of your circuit. you will probably need to add a cap to ac couple the rectifier circuit

>>1741077
the resistor on the left doesn't participate in the dc level on the right because caps block dc. select it to match what your load likes to see or just leave it 10k
on the right, the resistor is a Thévenin equivalent of one 20k resistor to ground, one 20k resistor to +5V (see Thévenin's theorem)

>> No.1741104

>>1740904
>load on emitter
>diode in wrong place
into the trash it goes

Look up what a freewheeling diode is supposed to do, but what's pictured isn't a freewheeling diode.

>>1741068
Yes but now I have no university stuff to do

>> No.1741110

Im playing with analog voltages and Im noticing jitter depending on what type of breadboard I use. Im moving the circuit to a solderable breadboard to eliminate that possibility. My question is can I still use the dupont cables and solder to them or should I just use solid core/ stranded wire.

>> No.1741115

>>1741110
> solder in dupont

I wouldn't. Just pull the telephone wire out of your walls if you're not using your land line any more and use that.

The jumper standard is bare tinned solid.

>> No.1741118

Do I need a high-voltage programmer to change fuses on a (DIP) ATMega328P?

>> No.1741128

>>1741118
does the datasheet have any notes about HVP-only for fuse setting in the programming section? if not, then it depends solely on whether LVP is disabled, no?

>>1741110
there are many ways to construct permanent versions of your circuits. dotboard certainly works
any conductive wire that's small and thin enough will do, but mechanical considerations come into play. avoid stranded unless you provide strain relief and (for longer runs) any other mechanical support to prevent them flapping in the breeze

>> No.1741131

do you guys have some go-to transistors for each type? Or do you look up the most optimal one for each project? I'm just wondering if I should buy like 20 of specific models just to have some inventory to be able to throw circuits together, or if i might as well just build it up project by project.

>> No.1741132
File: 809 KB, 4032x3024, image.jpg [View same] [iqdb] [saucenao] [google]
1741132

>>1740865
Tell me more. What is the plan? I still need to figure out what to do to into arduino stuff since that kit is collecting dust.

>> No.1741133

>>1741131
Those Amazon kits are so damn cheap, I always want to buy them. Of course the one you want isn’t included, but a healthy stockpile makes you feel like a champ.

>> No.1741138

>>1740703
before I add that polarized capacitor, it's supposed to be oriented that way, correct?

>> No.1741144

>>1741104
>>load on emitter
whats wrong with that

>> No.1741156

>>1741144
> what's wrong with load on emitter

It's not typical, and definitely not something you'd start with in transistor design.

>> No.1741159

>>1741131
>go-to transistors
yes. I *go to* my transistor stock box and use what's in there
those starter kits aren't bad *to get started*. they usually contain a variety of useful types in TO-92 packages, most of which are super cheap due to their relative popularity, in turn due to their usability and robustness. if you are very active in the art, over time you will add some other types to your collection which suit the specific work you're doing, e.g. low Rds(on) and/or Vgs(th) MOSFETs, higher-voltage BJTs, JFETs if you do a lot of audio stuff, higher-frequency RF transistors, various power transistors, or maybe you take the SMT pill and pick up some of the usual THT types packaged in SOT-23, SC-70, maybe dual transistors in SOT-23-6 packages)

>>1741138
yes, the dc voltage on the input jack is lower than the dc voltage at the voltage divider

>>1741144
the emitter is an input terminal and the output to a load on the emitter will vary according to the nature and behavior of the load. this is usually undesirable
https://en.wikipedia.org/wiki/Amplifier#Common_terminal

>> No.1741163

>>1741131
>>1741133

I suspect those kits are actually all one transistor (i.e. the 2N3904 is actually the same die inside as the 2N2222) and for most switching apps you won't notice a difference.

Also, 2N3906, MPS 6515, 2N5457, TIP31A, TIP32A, 2N3055.

That ought to cover it. ON semiconductor is reputable. Get metal can 2N2222s and 2N2905s.
Just add a bunch to your digikey/mouser order to get the free shipping.

>> No.1741165
File: 1 KB, 215x165, creating a virtual ground.png [View same] [iqdb] [saucenao] [google]
1741165

>>1741077

ahah, you shouldnt have listened to the anti-LM386 fags. the LM386 input is self-biasing, so you woulnt experienced any of your current problems.

anyway, to fix your shit, you gotta create a virtual ground, and reference all your ''ground'' signals to this instead of the negative of the battery. this automatically biases all your inputs and feedback paths to the half-way supply of your circuit so the signal has room to go both up and down, so you dont get signals with the top or bottom cut off, but sinewy beautiful full sine waves.

pic related is a good way to do it. use 10K for resistors. you can also eliminate the op-amp, and just use the centerpoint, but that's not as good. google ''virtual ground driver'' for info if you wanna read about it.

>> No.1741173

>>1741131
after one or two projects, and buying in packs of 10 units or more + salvaging shit you find in the dumpster you'll have a lot of random transistors you can use here are there.
>>1741144
You'll need a voltage on the base higher than the supply voltage to put it in saturation.
>pnp for highside switching
>npn for lowside

>> No.1741180

>>1741165
>just one cap
Theoretically just one cap is all you need, but I've heard people say otherwise.

>> No.1741183

>>1741077
ah sorry, I think you probably are using a single supply (positive - 0V) but the signal goes below 0. Opamps don't like that. You need to either add a DC component to your wave so it doesn't go past 0 or add a split supply (positive negative)

>> No.1741186

>>1741165
I'm still not 100% sure why using an lm386 is a bad idea other than "they're expensive and it's a waste"
I'm certainly having some rotten luck and a terrible go at it with this op-amp. Now I can't get the damn amplifier to amplify after I pass it the output from the voltage follower (a la >>1741084). It seems like the amplifier stage is just acting as a comparator and going high 100% of the time.

I know nothing about virtual grounds, and now I'm doubting if I really understand the term "biasing" so I guess I'm going to go hit the books

>>1741183
Yeah, it's an lm358 single supply op amp. I'm working on biasing the input now but I'm hitting some issues

>> No.1741217
File: 454 KB, 1645x1195, 1568735094940.jpg [View same] [iqdb] [saucenao] [google]
1741217

>>1741163
>the 2N3904 is actually the same die inside as the 2N2222
I very strongly doubt that your statement is true, Pic related
but I don't doubt that your statement is true for chink 2N3055s

>> No.1741234

>>1741186
consider the following: with a proper opamp to hand, you can build the circuit from Figure 22, Full-Wave Average Detector to VU Meter Specifications, verbatim. ac coupling is already there for you

>> No.1741235
File: 213 KB, 1062x1375, 1555988329516.png [View same] [iqdb] [saucenao] [google]
1741235

>>1741186
>>1741234

>> No.1741238

>>1741186
also don't forget the LM3916 has a VLO pin which you can tie to something other than ground to establish your zero-input point

>> No.1741249

>>1741217
no he was specifically talking about the bulk assorted transistors, and implying that they're all the same transistor with different labels etched into them

>> No.1741253

>>1741249
kek, oh, in that case he might be right. I am still inclined to demand curve tracer screenshots or it didn't happen

>> No.1741262

>>1741253
on that note, how do you curve-trace a transistor? Emitter to ground, resistor from collector to Vcc and a resistor on the base to the function generator, with Vbase and Vcollector as the X and Y axis? Or hooked up somehow such that Ibase and Icollector are X and Y instead?

>> No.1741287

>>1741262
depends on what curve you want to measure. since BJTs are current-mode devices, some sort of voltage<->current conversion would be involved in any interesting measurement
I'd probably try first to characterize the V-I curve of the b-e diode, since it's where most of the action is. both the forward and reverse characteristics would be of interest in fingerprinting the part type. then, transfer characteristics vs. Ic, feed a chosen current e.g. 10µA from b-e and measure the c-e current over a range of voltages, derive beta vs. Vc
the b-c junction alone is relatively loosely controlled and doesn't really participate as such in applications, therefore not too interesting. maybe there's a fun noise test in there somewhere

>> No.1741289

>>1741287
>not testing negistor efficacy

>> No.1741290

>>1741289
das rayciss

>> No.1741341

>>1741132
I am making myself a super cheap bench power supply, where i can use a rotary encoder to set a voltage and a maximum current using a display just like on commercial power supplies, except mine will cost like $15 total instead of $100. And because there is no point reinventing the wheel, then as a work horse of the supply i am using a buck converter that i already own and hacking into it's voltage and current control feedback loops

>> No.1741342

>>1741341
Cheapest PSU I could make would be an LM2596 that hooks up to an existing 12V/24V power brick, with a cheap voltage/current panel meter and a voltage and current pot. No digital feedback but I figure the drift would be low enough not to worry about. The actual case for such a project would cost more than the main buck board.

But personally I'd rather have a dual-rail PSU with an output that can be toggled as to whether its grounded. Not sure if any jellybean boards can give me a variable split-rail AND current limiting. Bonus points for somehow shoving a linear regulator in there for ripple-rejection to drop a constant volt or so.

>> No.1741346

what is the lm317 of opamps?

>> No.1741347

>>1741346
pretty funny actually

LM324

>> No.1741352

>>1741346
>>1741347
wait i use both of those, am i retarded?

>> No.1741353

>>1741352
not really, they are just very archaic 30 years old tech

>> No.1741358

>>1741352
they have their places
but RRIO op amps are cheap enough that they should probably be in any serious hobbyist's jellybean jar

>> No.1741364

>>1741353
>>1741358
>RRIO op amps are cheap enough
Yeah I guess so, I just fall back on the old parts because they're ubiquitous in chinkland and less likely to be fakes. Not too sure what parts I should replace them with as a general-purpose op-amp, there's too many of the damn things to pick from.

More importantly, what general-purpose adjustable linear regulator should I give a shot? I'm guessing these days you can get a few more dB RR and a few volts less minimum dropout, but I've heard that LDOs can get less ripple rejection when operating with a low dropout.

>> No.1741369

>>1741364
there are a few chinky LMV"321"/"358"/"324" replacements that claim RRIO but they are kinda hard to pin down. MCP6001/2/4 and LMV3xxTP from 3Peak might be bretty gud if 5V single supply is enough. if you look harder on LCSC you might be able to find RRIO with wider supply voltage but you might leave the realm of jellybean pricing
>general-purpose adjustable
no idea, I'm mostly a switcherfag, consider +5V my high voltage rail in the usual cases, and tend to use fixed LDOs and TL431-alikes on those occasions when I do need LDOs
the old add-a-tranny on the output of the opamp or TL431 trick still works great in 2019

>> No.1741370

>>1741186
I know the 358, I use them weekly. When they say "single supply opamp" it doesn't mean it can't run a negative supply. It means it can sense and give outputs that are very close to 0 (or the negative rail). What problems? Usually only a voltage divider and a capacitor should do it.

>> No.1741371

>>1741358
which ones?

>> No.1741517

What circuit design tool is best for android phones?

>> No.1741590

>>1741369
>TL431
If only they'd fucking arrive. I assume it wouldn't be too difficult to use them in both negative and positive adjustable regulator circuits.

If I slap a differential amplifier across the input and output of an adjustable regulator, and this as (duty-cycle) feedback to a buck converter powering the regulator to always give a constant dropout, would this make a good power supply? The regulator would be adjusted in the normal way, and it would be relatively easy to make this a split-rail since there are complimentary linear regulators.

Now that I look at the datasheet it's obvious that the LM317 is not suited for this task. Mainly its dropout of up to 2.5V, its poor ripple rejection near the MHz range, and also its minimum operating current I guess. I can't find any information on the minimum output voltage, but I'll assume it's equal to the 1.25V reference, which is perfectly fine for a split-rail power supply. For low voltages I'll use a different, single-rail power supply, likely with no linear regulator on the output, just a ∏ or T filter directly off an adjustable buck.

>> No.1741637
File: 599 KB, 1024x619, sva1032x-qam_ma.png [View same] [iqdb] [saucenao] [google]
1741637

I want to expand my knowledge as ES engineer with EE background to knowing how to use a spectrum analyzer / vna in order to tune antennas and design filters.
Do you guys think the Siglent sva1032x spectrum analyzer vna combo is a good investment right now? I have some money saved up but I still have to finish my thesis. However, I think it would be a very good addition to my skillset and an awesome device to have.

appearently, its about 3.6k euro incl tax, excluding some options
https://www.siglent.eu/siglent-sva1032x-spectrum-vector-network-analyzer.html

>>1741517
I prefer Altium but I hear KiCad is great too these days. Also will you be ok designing a phone if you are not really oriented in pcb design tools?

>> No.1741664

>>1741637
>I prefer Altium but I hear KiCad is great too these days. Also will you be ok designing a phone if you are not really oriented in pcb design tools?

Sorry for the confusion, I mean android apps for circuit designing

>> No.1741698
File: 304 KB, 1920x1440, 1547495286031.jpg [View same] [iqdb] [saucenao] [google]
1741698

>>1741517
just use falstad

>>1741590
interesting idea, to use the voltage across the pass device as feedback to another regulator. sounds like it might be hard to compensate and prone to self-destruct. KISS

>>1741637
I wouldn't blow €3600 on test equipment unless I was using it more or less full time to amke money
you could learn the concepts with one of the OSHW VNA projects and an RTLSDR for 1.5 orders of magnitude cheaper
>ES engineer

>> No.1741713
File: 136 KB, 1093x1500, scoot.jpg [View same] [iqdb] [saucenao] [google]
1741713

>find pic related on the side of the road for free
>dirty as fuck but at least it rolls
>doesn't come with charger
>it uses lead-acid batteries and a chain-driven motor
>test batteries, 24 volts
>motor moves when hot-wired to batteries
>hook everything back up
>hear the relay in the motor controller clicking on and off when I turn the throttle
>controller still reacts to throttle when the 3-pronged on/off switch is set to "off"
>scooter doesn't move, even when pushing it along like the manual would suggest
>hear the relay clicking on and off while holding the throttle down

I can see a few possibilities:
1. The batteries are actually drained/fucked and I didn't realize it ($12 for the charger, $25-$35 for the batteries)
2. The throttle is fucked ($15)
3. The controller is fucked ($15)
4. I wired something up wrong (Priceless)

So I have a few options
1. Fix it with stock parts
2. Replace the internals with something better
3. Gut it and use it as a kick scooter with a storage compartment

What do you guys think?

>> No.1741745

>>1741346

LM358. They're still making new ones (they just released some). TI calls it "industry standard" and they're not wrong.

http://www.ti.com/lit/ds/symlink/lm358ba.pdf

>> No.1741768

>>1741713
I'm guessing water ingress into the controller circuit. Take the housing apart and get a good look at the control circuit. If it's a brushed motor I'd pick option 3, otherwise 1 or maybe 2 with a cheap lipo pack.

>> No.1741922
File: 26 KB, 640x354, hoverboard-fire-screencap.jpg [View same] [iqdb] [saucenao] [google]
1741922

>>1741768
>cheap lipo pack.

great idea if you wanna renovate your house with the insurance money after a fire.

>> No.1741932

>>1741922
just don't buy chink shit

>> No.1741934

I have led bulbs which support dimming but turns out ssr isn't fast enough to dim them when i trigger it with pwm since it has up to 10ms reaction time. Shutsucks man.
What do i do? I don't know of any other way to trigger 220V fast enough where it can handle couple of amps

>> No.1741941
File: 6 KB, 336x150, triac with MOC3010.jpg [View same] [iqdb] [saucenao] [google]
1741941

>>1741934

triac with MOC3010

>> No.1741942

>>1741941
is it physically isolated with logic level support? i need to control it with a 5V micro

>> No.1741943

>>1741942
>is it physically isolated with logic level support? i need to control it with a 5V micro

yes and yes

>> No.1741947
File: 4 KB, 449x93, file.png [View same] [iqdb] [saucenao] [google]
1741947

>>1741943
Nice.
They have a bunch of variations, is 3010 the best one?

>> No.1741951

>>1741947
scratch that, you need MOC305x for grownup mains

>> No.1741953
File: 268 KB, 774x962, 1548613093974.jpg [View same] [iqdb] [saucenao] [google]
1741953

>>1741932
>LipO bAtTeRiEs aRe tOtAlLy iNerT aND jUsT tOyS

>> No.1741955

>>1741953
why do you hate on fire? it keeps you warm and safe from predators

>> No.1741965

>>1741951
Okay thanks. 800ma forward current will be plenty.

One more question, in the datasheet i can't find the maximum switching speed, how quickly does it react?

>> No.1741980

>>1741953
did you know that also electricity can kill you?

>> No.1741984

>>1741953
>NOOOO YOUR DIY PROJECT HAS TO MEET SAFETY STANDARDS YOU HAVE TO GET IT CERTIFIED
getting zapped by mains or blowing up a battery is a great way to learn electronics.

>> No.1742093

>>1741698
Well in a way I am using it to make money. Currently, I am using a third party to match my antennas at work, but I'm quite confident I could do this myself with the right tools and with the knowledge of the EE bachelor. Maybe I'll just let my boss buy one and if I like it I'll buy one for myself too later, that would make it at least way less risky.
To be honest, I wouldn't really like any oshw in my lab. last time I tried oshw was with daves ucurrent, and then with the current ranger, but both suck and I could do it better myself. would've been better off just buying a bench top power analyzer or something.
Also, why did you feel the need to quote me on ES engineer?

>> No.1742095

>>1741664
I don't know any, but I dont think anything useful exists.

>> No.1742144
File: 108 KB, 2000x1751, 1551901282900.png [View same] [iqdb] [saucenao] [google]
1742144

>>1741965
On Semi's datasheet on p6, "LED Trigger Current vs. pulse width", hints that it's kinda slow. the device is specified for a 6-15mA at 100µs trigger pulse (minimum), with conversion factors in case you want to run it shorter. but you're just chopping a 10ms half-sine so I don't think you need super great precision anyway

>>1741980
I heard the same thing about dihydrogen monoxide. fake news

>>1741984
causing a fire that one is probably ill-equipped to extinguish is a bit more of a lesson than babbies bargain for, I think

>>1742093
>could do it better myself
well that's fair. OSHW is only free if your time has no value
>ES engineer
we like giving people shit around here, that's all

>> No.1742268

Any recs for a hot air station? Currently using a cheap Weller station for soldering. I'd like to upgrade that and get a hot air station; Are the 2-in-1 things any good or should I just buy a T12 and a hot air station separately?

>> No.1742274
File: 147 KB, 1000x1000, 1554706194130.jpg [View same] [iqdb] [saucenao] [google]
1742274

>>1742268
2-in-1s aren't inherently any worse than the separate units. decide based on your own space, transportability, and free outlets situations
I've been fairly happy with picrel and I proudly shill picrel

>> No.1742281

How easy would it be to construct a nitrogen (or CO2) flow from a soldering iron?

>> No.1742282
File: 27 KB, 930x582, 1575772641227.jpg [View same] [iqdb] [saucenao] [google]
1742282

>>1742281
as hard as you like

>> No.1742289
File: 309 KB, 1000x1000, 968-fume-extractor-handle.jpg [View same] [iqdb] [saucenao] [google]
1742289

>>1742281
or as easy as you like

>> No.1742292

>>1742282
I don't suppose anyone here's ever soldered with one of those?
Now that I look at it, my T12 station's pencil has a small gap around where the tip comes out, and the tip only protrudes ~50mm from the grip, so I could probably just pipe gas into the inside of it and have it spray out in a fairly orderly fashion, especially if I use a tad more gas than the professional models. I typically go full over-flux when soldering, but perhaps with nitrogen I could go really light on the flux (and/or only use really inactive flux), which could be preferable for SMD stuff where getting flux under a QFP would spell trouble. I'd give it a try if I had an arc welder (for the inert gas) and a spare handle.

>>1742274
>>1742289
How universal are those plugs? That 986 one especially looks really similar to my T12 one, and if they have the same thermocouple+element circuit and run off the same voltage it would be neat to be able to swap them. Because the T12 is a good, cheap, customisable station, but the tips are more expensive than those dollar-a-dozen copper tips on ebay.

>> No.1742305

>>1742292
>I don't suppose anyone here's ever soldered with one of those?
I fucking wish
>How universal are those plugs?
plugs seem to differ between makes or maybe even models with the 900-M tip system. the circuitry seems to be about the same: two leads for heater, two leads for TC, one lead for ground
otoh, the JBC-handle-on-chink-T12-controller has been done and it's uma delicia https://www.youtube.com/watch?v=g3d1rZa-i5M

>> No.1742307

>>1742305
I thought the thermocouple was in series with the element on a T12? It only has 3 pins after all, but maybe their thermocouple reading ICs are fine with either configuration since they probably use differential (instrumentation) amplifiers.

>> No.1742356

>>1742144
>a 10ms half-sine so I don't think you need super great precision anyway
wait i just realized, since triac is zero crossing triggered and the mains frequency is 50hz, then i will only get 50 on or off states per second which is also why my ssr has 10ms reaction times so that won't solve anything because i bought dimmable 230v leds and using pwm where one pulse is say 50ms long just doesn't work.
I wonder how commercial led dimmers do it

>> No.1742357
File: 740 KB, 3252x867, IMG_1964.jpg [View same] [iqdb] [saucenao] [google]
1742357

>open up my old soldering iron with the analog temperature dial on the side
>has four wires going into the tip
Wait does this actually have a thermocouple or something in the tip? You guys told me all of these had TRIAC dimmers instead of proper temperature feedback. The main IC says:
>358
>GZS46
With an ST logo on it, so I'm guessing it's an LM358 dual op-amp. I'm interested to see the circuit, as it runs directly off mains. There's a power-resistor near the op-amp and what could be a zener, so I assume it's some sort of DC supply.

>> No.1742359

>>1742307
oh, thought you were talking about the 900-M. no idea about T12

>>1742356
same as any other phase angle dimmer. triacs can't turn off until zero current through them, so all you can do is choose when in the half-cycle to turn them on. that means you need to know when the zero crossing happens which probably means another opto in the other direction

>>1742357
neat, that means it's somewhat repeatable even if not very accurate or precise. once you have stabilized the tip temp using some external gear you can simply tape over the handle and let your dumb techs use it

>> No.1742363
File: 496 KB, 3257x695, IMG_1963.jpg [View same] [iqdb] [saucenao] [google]
1742363

>>1742357
Oh this is surprisingly difficult to trace out, I'll give it a shot later when there's better light. Here's the other end just for interest. That diode D3 is a normal rectifier and that unnamed diode with a blue stripe is connected to it in antiseries as a zener. R5 and R6 make a divider from live to neutral, presumably for peak detection or something along those lines. The TO-92 is a MAC97A6 TRIAC, so yes it likely is for timing purposes. The blue wire goes straight to the IC through a 100Ω resistor, while one of the two white wires is being switched by the TRIAC while the other is tied to live. The red wire is tied directly to neutral. The white wires have a resistor symbol at their root while the red and blue have what looks like a thermocouple symbol at their root, which is rather promising.

>>1742359
Yeah, as far as a beginner iron goes it's basically perfect. It doesn't need to be precise at all, it costs $6 on alibay, and it avoids thermal overshoot and shouldn't be underpowered.
It looks nearly identical to this one:
https://www.aliexpress.com/item/4000118479529.html
But I'm unsure if these cheaper ones have the same circuitry:
https://www.aliexpress.com/item/4000189516230.html

>> No.1742384

>>1742359
>so all you can do is choose when in the half-cycle to turn them on
ohhh, is that why my shit wasn't working? because my PWM wasn't synchronized with the zero crossings in the mains ac cycle?
could you please provide a simple example circuit of how i can detect the zero crossing with my 5V DC microchip?
I imagine i need some way of monitoring the ac voltage really fast and constantly check if the voltage between live and neutral is 0? but i am not sure how to do that

>> No.1742390

>>1742384
wait i had a big brain moment,
if i take an octocoupler and connect the 220V to the LED part of it as in
resistor -> 1x diode (or 4x diode to form a full bridge rectifier) -> octocoupler led
and then connect my micro digital pin to the other side of the octo, then every time the octocoupler is off i know the zero crossing is happening
would that work?

>> No.1742392
File: 9 KB, 400x400, tegaki.png [View same] [iqdb] [saucenao] [google]
1742392

>>1742384
Normally this is done with a DIAC, and I think you could maybe still use a DIAC in an electrically-variable PWM circuit with resistor biasing or something, but I'd go for something more like a standard comparator-based PWM system. Generate a sawtooth wave (just an integrator?) from a zero-crossing detector (detector can be on the mains side with the sawtooth gen on the other side of an opto). Then feed the sawtooth in one side of a comparator and a variable DC voltage in the other side. Feed the output of the comparator into another opto and then into the TRIAC. If you've done this correctly, a potentiometer or other variable voltage source will cause a voltage that corresponds directly to the duty-cycle. If you want an MCU to choose the duty-cycle itself, you'd need to output a high-frequency PWM, and low-pass filter it into an approximate DC voltage. If you do just want to use a potentiometer for dimming, use a normal TRIAC circuit.

The more synchronous options would be to somehow have the MCU running on a multiple of the mains frequency (possible with a PLL) or else using an interrupt each zero-crossing, and a simple delay before activating the TRIAC. Not sure whether the continuous interrupts will cause havoc to other parts of your circuit, I know some peripherals won't like having other interrupts in your code.

>>1742390
Yes, I think that's a semi-common 0-crossing method. At the very least I'm pretty sure I've seen something like it, perhaps with two antiparallel LEDs heat-shrunk next to LDRs in an audio circuit or something.

>> No.1742415
File: 1.42 MB, 1073x1804, 6mi7cp0fgbky.png [View same] [iqdb] [saucenao] [google]
1742415

Any VX junkies in here?

>> No.1742417

>>1739171
Noice

>> No.1742427
File: 198 KB, 970x728, 1674-02[1].jpg [View same] [iqdb] [saucenao] [google]
1742427

Anyone want to wager a guess as to what gauge this wire is? The product page doesn't list it, and I want to buy a spool of wire roughly close to its size to use with it.

>> No.1742441

>>1742390
that is correct. there are optocouplers with antiparallel LEDs that are designed for just this thing. SFH620A is a not uncommon example
ignore the other faggot, you don't need analog complications for this if you just want to git r done

>>1742427
22-20awg at a guess

>> No.1742464
File: 343 KB, 2048x1536, IMG_20191230_153600.jpg [View same] [iqdb] [saucenao] [google]
1742464

>>1742392
thanks fren, i purchased the parts required for the simple fbr solution so wish me luck, i hate working with mains because they are scary, one mistake and you get darwined

i got reminded why i shop exclusively in chink shops. i didn't wanted to wait two months for the parts so i went to the local radioshack and bought the parts there. For the price of a single optotriac there, i could buy 10 of the exactly same ones from ali

Fucking greedy bastards.

>> No.1742478

>>1742464
>i hate working with mains because they are scary, one mistake and you get darwined
know what you are doing and have confidence in yourself. also keep one hand in your pocket (literally) whenever you have any exposed mains circuitry powered up. then, you can be fairly sure that if you get poked at least you won't get fibrillated
>greedy
to be fair, warehouse space ain't cheap if you're not turning it over on a daily basis
but yes, Digi-Key and Mouser are jews

>> No.1742479

>>1742441
>SFH620A
Thats neat, so you don't need any rectification at all, only one resistor with that?

>> No.1742481

>>1742478
>Digi-Key
I don't understand how they are still in business
one arduino uno costs $22 there, while chinks sell them for $3 and they aren't like a shit knockoff either, i have 10 of them and they work flawlessly. that is 7!!! for the price of one

>> No.1742487

>>1742479
two: one for series with the LED and one to pull up the phototransistor. when the line is around zero volts the opto turns off and the phototransistor stops conducting
note that you won't see the exact zero, of course, you'll just see when the voltage is close-ish. don't make the series resistor too small or you might be more susceptible to noise. for best accuracy you will probably want to to time the length of each zero detect with the hardware timer and offset your turn-on pulses by half that interval, or just fudge it

>>1742481
>shit knockoff
well they're not official Arduino-licensed merch either. so whether or not they're shit, they are knockoffs
I think it's people who build smallish volumes with unusual parts and don't keep a lot of their own stock that keep Digi-Key in business. stock your basic passives from LCSC

>> No.1742494
File: 141 KB, 1074x538, please_kill_me....jpg [View same] [iqdb] [saucenao] [google]
1742494

How critical are traces of differential pairs for USB 2.0 devices? Will my circuit fuck up if they're not perfectly even?

I'm connecting three microcontroller through a cheap chinese USB IC and I need to connect it all, but I've never done USB stuff just very basic digital PCBs with Atmega chips.

>> No.1742504

>>1742494
12Mbps will let you get away with a lot but not 480Mbps
the lack of a consistent ground plane for currents to return on means distortion and noise on your USB signals, and also higher EMI emissions as the electrons hit speedbumps and fly off the trace. you should really keep a consistent, solid ground plane under your impedance-controlled traces. consider a 4-layer board if you can't route around that, so to speak
use the pcb calculator to figure out the correct width/spacing for your traces according to your board material, copper layer thickness/stackup, etc.
use the differential routing tool to enforce that width/spacing and keep each arm of the diff pair of equal length
check out the app notes and layout considerations of western IC makers' data sheets for USB stuff. for example https://e2echina.ti.com/cfs-file/__key/telligent-evolution-components-attachments/13-106-00-00-00-00-33-10/USB-2.0-Board-Design-and-Layout-Guidelines.pdf

>> No.1742506

>>1742504

Thanks. I'm learning and doing all at home for fun so I'm not going past cheap 2-layer boards.

As for the planes I always fill at the end both sides with GND but I've read other people do one side for GND and the other for VCC, does it matter much?

>> No.1742514
File: 109 KB, 1200x1200, preview.jpg [View same] [iqdb] [saucenao] [google]
1742514

I have this electronic guitar and the 1st chord (e) is lower volume than others. Did you guys meet with this before?

>> No.1742517
File: 103 KB, 864x900, 20191230_115123.jpg [View same] [iqdb] [saucenao] [google]
1742517

I've seen EEEVblog's video on the cockcroft voltage multiplier he says it's not really made to drive loads but mostly used for electrostatic applications which got me thinking, couldn't I just put multiple high farads caps in parrallel at each capacitive junctions to increase its current output?
In the video he uses some very low capacity film caps to accentuate the voltage drop at Mega-Ohm loads.
Also, how do Liquid-Caps fare in those?
I'm aware this is a pretty easy way to kms if I'm being a retard.

>> No.1742519

>>1742517

bigger caps means more current, for sure. problem is cost. big caps at high volts are very expensive coz very rare.

>> No.1742523

>>1742519
He mentions you don't need higher voltage Caps though.
The only potential difference that matters is the one between each section, which is almost the same as the source, minus ESR & Diodes.
You can see him raise the voltage to 80v in 20v increments using 50v caps
Probly still cheaper to use a chunk of copper and iron to raise the voltage.

>> No.1742539
File: 238 KB, 354x349, 555.png [View same] [iqdb] [saucenao] [google]
1742539

What brand is this?
I ordered TI ones but I think I got ripped off.

>> No.1742542

>>1742539
Looks like Signetics.
555's are so common and similar from producer to producer it doesn't really matter unless it's fake.

>> No.1742552

how would i go about finding when the bottom of a voltage ripple occurs? Differential op amp?

>> No.1742559

>>1742552

a comparator will do. adjust it down with a 10-turn pot until it just starts to put pulses on the output. measure voltage at center of pot.

>> No.1742618

>>1742464
>>1742478
Yeah if it's only one hand on the circuit you'll be fine. Even 240V mains across my finger and my heart felt fine. Finger sustained a burn, but otherwise fine.

>>1742514
You get that issue on both pickups? Is it a function of what string you're plucking or on what frequency you pluck it at?

>>1742539
>555 timer
you got ripped off alright.

>> No.1742627

>>1742618
every online store only had SMD ones so I just bought these for 1€ which I honestly regret
and because of my paranoia I didnt want to buy the cheaper NE ICs

>> No.1742662
File: 28 KB, 773x356, level.png [View same] [iqdb] [saucenao] [google]
1742662

>>1742618
It's the same on both pickups. It's an ibanez gio, value oriented model. Could this be the issue? I just redid most of the solder joints but the issue persists. The bridge wasn't grounded well, I was getting around 150 ohms but I scratched some paint off where the ground was making contact under it and now i'm getting 0.5 ohms. Also the pickups were wired from the factory as follow: The top one's signal was going to the switch and the ground to ground; The bottom one was reversed, the signal to ground blob on volume knob and the ground to the switch. The switch is already grounded. Why were they wired like this? I changed them so both the signals go into the switch.

>> No.1742664

my building's parking lot has a bunch of abandoned cars and i'm super tempted to just go in and get my hands on their alternators and windshield wiper motors

>> No.1742699

>>1742481
arduinos are open source hardware.. Honestly I don't get why people complain about aliexpress.
The only problem I had was with the brazilian mail "not finding my adress" in the middle of a busy capital, my guess is the albel must have been accidentaly somehow. I mean, most ICs are already made in china (how many fabs are outside of china/korea?? 5?) and there is this big fear of knockoffs but I always get my refunds from ali.

>> No.1742700

>>1742699
And most chink sellers are like 100% supportive (sure in broken english and criptic messages, but they always fucking try)

>> No.1742702

>>1742699
Agreed, and if you're not gonna use aliexpress, Lcsc is pretty great.

>> No.1742744

>>1742506
two different functions. as you know, current travels in circuits. the plane serves as the return for signal currents. it is necessary to make the *area* of the loop for high-speed signals as small as possible to minimize distortion and inbound/outbound EMI. therefore it is usual to route them over a continuous plane if at all possible
>does it matter much?
from a high-speed signal perspective, you can see from the above how it does. there are tricks e.g. cap-coupling the planes together at the point of discontinuity
from a general perspective, I usually fill both sides with ground and carve out higher-priority power polygon(s) for other rail(s)

>>1742539
that N is NatSemi's (now a TI company, lel). Signetics was squarer

>>1742627
the large type boomerchips are drying up. SOIC-8 is ezpz, use it and love it while you can

>>1742662
>the top one
>bottom one
filthy casual. that's neck resp. bridge position
>why
basic physics, my dude. each pickup receives harmonics of the played string in different measure, and in different phase, according to the pickup's position along the string. with the bridge pickup reversed, and assuming it's wound all the same as the other one, notes played at the 12th fret and higher will be emphasized in their even harmonics https://newt.phys.unsw.edu.au/jw/strings.html
no shame in adding a reversing switch for it, if you want to hear the difference

>>1742699
they're bretty gud for module boards "for arduino", cheap hand/soldering tools, and stuff like that. but man, when LCSC will cut off a strip of 100 chip resistors for only 4 cents, it's hard to justify the whole flea market thing and not being able to have some python script write up a BOM to stock your lab
>how many fabs are outside of china/korea?? 5?
pretty sure Microchip alone has 5 in Arizona and Texas
>fear
it's not necessarily being out the money that's the killer. you still spent 15 macaques and two months getting that envelope into the country, right?

>> No.1742772

Holy shit you can get TO-92 and SOT23 PROMs, EPROMs, and EEPROMs. Though the EPROM datasheets say OTP-EPROM, which I can't imagine being any different from ordinary PROM. There's also something called Add-Only Memory, which isn't something I can find a wikipedia page so.
They're mostly 1024 bits or so, and I think they're used for storing some sort of address. They all seem to use "1 wire" as their communication method of choice.
https://datasheet.lcsc.com/szlcsc/Maxim-Integrated-DS2431-T-R_C295056.pdf
This one uses a built-in CRC, which is interesting.

I just find it strange that we're using a rather high-level digital communication method to a TO-92 of all packages.

>> No.1742773

while we're on the topic of guitars, and their pickups, i have a question:
so when a string vibrates up-down-up-down, it goes closer and further from the pickup, hence making a signal at the fundamental frequency
but when a string vibrates from side to side, it gets closer and further from the pickup twice every vibration, so wouldn't it make a signal at twice the fundamental?

>> No.1742786

>>1742772
huh. acorn archimedes computers kept their unique hardware IDs in a socketed to-92 rom. thought it was a dead technology.

>> No.1742800

>>1742786
Oh that sounds neat. I'd like to see these being used by hobbyists more often
https://datasheet.lcsc.com/szlcsc/1912101819_Microchip-Tech-11AA010-I-TO_C461607.pdf
This one goes up to 16kb and is only ~20c each. 16kb is more than enough for a password or private key, you could store a bunch of personal presets or URLs or hashes or whatever.

Is using 1-wire TO-92 EEPROMs more esoteric than CDIP quartz-window EPROMs in what's basically the 2020s already? I'm still leaning over towards the EPROMs because they can be used in place of combinatorial logic (and probably quite a lot of raw text if I wanted to make some sort of primitive reading device).

>> No.1742819

>>1742773
theoretically, but a wound string's vibrations would start to go circular, I'd think

>>1742772
"1"-wire has been around for a long time, I remember screwing around with Dallas Semi's 8051-based deeply-integrated Java micro, which was too cute by half. they tried to market it in building automation and security token markets, with some success
I believe those TO-92 devices are active SKUs as much out of inertia and government procurement as anything
>add-only memory
just a shim of constraint logic between you and the EPROM bit cells

>>1742800
anachronistic ≠ esoteric

>> No.1742854

>>1742819
>circular
oh so you'd get both harmonics

>> No.1742862

>>1742854
perhaps not from the very first cycle, but I would SWAG that changes in tension along a wound string would cause torsional forces and the vibration would eventually tend toward circular polarization

>> No.1742870

>>1742862
well harmonics do change as time progresses. iirc they get less harsh, which kinda makes sense if strings are normally plucked side-to-side and the circular harmonics would be more fundamental and less of the second
but that's just folk-physics

>> No.1742872

>>1742772
Not new, this was from Dallas Semi before Maxim bought them out, I wrote assembly language to bit bang it in 1998 or so, back when you could still count cycles reliably. We were just using it as a board serial number chip.

>> No.1742890

>>1742872
>back when you could still count cycles reliably
Does that refer to the increase of clock speeds?

>> No.1742896

>>1742890
pipelines and caching and variable memory access speeds mean you can't count exact cycles anymore

>> No.1742972

>>1742744
>they're bretty gud for module boards "for arduino"
never bought those. Only regular parts, ICs, transducers, bearings, rails, motors and sometimes ordered a PCB there because of a promotion. I'm saying it's a silly fear because of what I read here
>hurr durr I don't buy timer 555 from china because I might be buying fakes
What does that even mean?
>I bOuGhT a ThOuSaNd 3 PiN IcS aNd 5 Of ThEm DoN't WoRk, I wAnT mY mOnEy BaCk
ok

>> No.1742975

>>1742972
>What does that even mean?
In the sense of how do you think that what is wrong with what you are doing is not your circuit/implementation but the fake of one of cheapest, most fabricated ICs ever?

I understand buying a 1000uF cap and getting a empty steel can with a 1uF cap inside, or buying a expensive PIC and getting a cheap one instead. But saying shit like counterfeit 555s and fake transistors is like believing in boomer-like facebook viral plastic egg posts.

>> No.1742983

>>1742890
modern processors and systems are less deterministic these days, containing lots of separate clock domains, speculative or lazy algorithms, multiple bus bridges, and other such fun. you can see a lot of this even at the lowish end of ARM MCUs

>>1742972
oh, that is true, their light mechanical part selection is also respectable

>>1742975
exactly. shit like plain old DIP 555s are so not worth faking. marginally-working pulls are a somewhat more possible possibility tho, and I wouldn't put it past them to blacktop a bunch of bipolar 555s as CMOS

>> No.1743001

>>1742975
>>1742972
To end this, the only real problem I had was when I had to buy lamps and power supply. (Both growlamps for greenhouses and normal highpower leds). You have to always expect 30% less power/current and a worse power factor than advertised and depending on how cheap is what you are buying EMI and electrical protection stuff may not be up to any code.

>> No.1743002

>>1743001
>30% less
oh
well shid
I bought a big bag of (35) Hong Li emitters a few months back and luminous efficacy is kind of important for horticulture. how fucked am I?
not even worried about drivers, I'm doing my own

>> No.1743027

>>1737362
Hey. I'm a CS&E grad trying to learn hobby electronics, mostly using microcontrollers.
I learned about circuits in physics and designed simple adders, counters and stuff in digital design.

I got
>Mims III, Getting Started in Electronics
>Platt, Make: Electronics
from libgen and I'm currently working through them. However I want to be able to design my own circuits and do the math and shit and so far it doesn't seem like I'm learning any of that. I'm also going to go through this class: https://www.coursera.org/learn/linear-circuits-dcanalysis

Anyway, how do I move from starter projects to designing my own circuits? I'd like to make mostly smart home/iot things, as well as cheap small robots (took a robotics class and I specialize in machine learning now so I'd like to fuck with real world robots)

>> No.1743031

>>1743027
fortunately you can do a lot of things without much math. the math you need is typically simple algebra for v=ir, opamps, passive dividers/filters, stored energy, or tolerance analysis. for your area of interest you probably care more about learning about sensors, protocols and power supplies so you can plug modules together.

if you want to learn some math instead then design your own bench power supply.

>> No.1743036

there are so many capacitor types bros... how do you choose?

>> No.1743039

>>1743031
Thanks. I was actually planning on doing that with a power supply from an old PC. Was thinking I'd use a fan from it to make a soldering fan using a filter like the one on the stovetop thing that sucks air.
I don't care about doing math as I have plenty of that already, but I'd like to learn how to design and analyze circuits.

BTW is someone going to make a new thread?

>> No.1743043

>>1743039
*use a fan from the old PC, not the power supply

>> No.1743044

>>1743027
there's not really that much math in the mechatronics/MCU end of things. it's all pretty cookbook unless you want to do some pretty serious signal processing in the code. that is, you take one fresh chip, stuff with decoupling capacitors, add your choice of crystal (optional), prepare a side of power supply, place in an oven at the standard JEDEC profile for about 4 minutes, let cool, garnish with sensors and switches, program and serve
those books are kinda shit for someone on your level. you should get The Art of Electronics instead which treats you like an adult. Platt's about to into the trash for stupid shit that you can read all about upthread

>>1743039
baking

>> No.1743047

>>1743044
Thanks fren. Going to check out the art of electronics. Platt kind of threw me off with the "put a battery in your mouth to learn how electricity feels" bit at the start.

>> No.1743048

>>1743046
>>1743046
>>1743046
>>1743046
>>1743043

>> No.1743592

>>1742744
>>1742662
I was thinking it had to do something with the waves. My mind flew to speaker wiring, thought I had to put them in sync.

>> No.1743593

>>1743592
What about the low volume on 1st string? Are the pickups of very low quality?