/ohm/ - electronics general, well-stocked workbench edition

[Anonymous No.1503466]

trace of old thread: >>1498817

>RULES
0. Electrics ≠ electronics. Mains wiring goes to /qtddtot/ or /sqt/. PC assembly to >>>/g/.
1. Do your own homework. Re-read all documentation/datasheets related to your components/circuits, and do an honest web search, before asking.
2. Pics > 1000 words. Post schematic/picture/sketch/9001.5 hours in MS Paint with all part numbers/values/etc. when asking for help. Focus/lighting counts.
3. Read posts fully. Solve more problems than you create.

>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
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, especially good for component assortments/sample kits (caveat emptor)
Your local independent retail electronics distributors
ladyada.net/library/procure/hobbyist.html

>Related YouTube channels:
mjlorton
paceworldwide
jkgamm041
eevblog
EcProjects
greatscottlab
AfroTechMods
Photonvids
sdgelectronics
TheSignalPathBlog
BigClive

>Li+/LiPo batteries
Read this exemplary resource first: https://www.robotshop.com/media/files/pdf/hyperion-g5-50c-3s-1100mah-lipo-battery-User-Guide.pdf
>I have junk, what do?
Take it to the recycler.

[Anonymous No.1503468]

>>1503466
this thread's digits brought to you in part by the LM3466 multi-string LED current balancer

[Anonymous No.1503503]

Alright i decided to use AA liions, 9900mah capacity is a shitton AND they are rechargeable too. this is great.
chinks sell them pretty cheap too, i just ordered a 20pack.
thank god for chinks, i wouldn't be able to afford a fraction of the nice things i have if i bought them here with insane mark ups

[Anonymous No.1503505]

>>1503466
>workbench with no space underneath for your feet or legs for when you want to sit at the bench and work
retarded benches like this. why?

[Anonymous No.1503510]

>>1503505
It's a standing station. It's much healthier than sitting on your fat ass all day growing blood clots.

[Anonymous No.1503517]

>>1503503
900mAh is pretty impressive for a 14500. 9900mAh is horsefeathers.

[Anonymous No.1503521]

>>1503510
Wow great so when I can't actually make anything because small intricate parts are too far away from my eyes to see and break my back bending over to steady my arms on the bench and shit and don't actually get any work done to pay bills and end up fucking homeless at least I'll have my health except the previously mentioned back problems.
OP is clearly too short for standing at unless its for a midget

[Anonymous No.1503538]

>>1503521
Insensitive asshole.
That bench was made for a Vietnam war veteran in a wheelchair who got his legs blown off by a faulty lipo battery in a Vietnamese radio.

[Anonymous No.1503541]

>>1503538
F

[Anonymous No.1503555]

>>1503505
Most workbenches are like that....

You're thinking of a desk or something.

[Anonymous No.1503566]

>>1503517
it's 18650

[Anonymous No.1503575]

>>1503566
AA sized are 14500, 14mm diameter by 50mm length, and the final zero is because fuck-you-we're-adding-a-zero-at-the-end. Even 18650s, the highest capacity ones are around 4,000mAh.

[Anonymous No.1503579]

>>1503575
The how do you explain this?
You are making some very serious accusations here.

[Anonymous No.1503582]

>>1503579
those clever chinese must just be lightyears ahead of everyone else on battery energy density

[Anonymous No.1503583]

>>1503582
you are reading it upside down
it's actually 0066mah

[Beppu !n/ZnkB9jUw No.1503586]

>>1503579
w2c?

>>1503466
I could have a workbench twice that size and I still wouldn’t have enough space to work because it would be covered with a bunch of half finished projects.

>tfw power supply is here but I can’t open it for another month
So many toys sitting here that I want to play with already but can’t. Did I do gud with the linear supply? I know it’s only one output and nothing fancy, but what are you gonna get for <$100.

[Anonymous No.1503588]

>>1503579

[Beppu !n/ZnkB9jUw No.1503591]

>>1503588
I think those are fakes. Too good to be true. But I heard that pic related is a good deal at <$2ea.

[Anonymous No.1503615]

>>1503579
>The how do you explain this?
The same way you explain 5000 watt car stereo for $99.99.
>it's bullshit

They're chinks. They don't have to abide by the advertising rules of your country.

[Anonymous No.1503624]

>>1503294
Why not use the original circuit? 1 in 1000 is more than enough resolution in any case, and at a much faster sample rate than 800/50 = 16 samples per cycle = 8 samples per half-cycle, which I'd say is pretty borderline as far as getting accurate data. I'd prefer at least 20 samples per half-cycle.

>>1503422
Can confirm, digisparks are the shit. And if that's too big for you, get yourself a handful of PDIP-8 ATtiny85s and a tiny USB programmer.

[Anonymous No.1503636]

>>1503624
>>1503483
>>1503483
thanks anon, I wanted something more diy, homemade. I`ll keep looking.

[Anonymous No.1503660]

>>1503468
Not gonna lie, I would like to see this every new thread (btw I don't come here often so idk how long this has been going on)

[Anonymous No.1503689]

>>1503660
For a while. But like half the time they're LED drivers or switching supply controllers.

[Anonymous No.1503770]

When a scope is isolated, against what ground reference it measures voltage?

[Beppu !n/ZnkB9jUw No.1503771]

My friends... I have another retarded question-

So I wired these 4 little panels in series. Maybe they will put out nearly 20v with good sunlight, but it’s 3am so my best bet is getting 5V from the flourescent lighting, 9V if I hit it with this flashlight.

Anyway if I were to hook them up directly to a Li-Ion battery and let’s say with the 4 panels in parallel I got 5.0V with 40mA, or I run them in series and it ends up being 20.0V but only 10mA, is it all the same in the end? Would both charge the ~3.7 battery the same? Or did I misunderstand something?

[Anonymous No.1503772]

>>1503770
The negative probe; the probe's ground clip. If you've been using your scope without a ground clip all this time I'd be surprised if you were getting optimum results, for interference reasons.

>>1503771
>hook them up directly to a Li-Ion battery
You'd see 3.2-4.2V across the cell because that's the voltage of the cell. If there's a resistor in series with the cell then you'll get whatever the internal equivalent circuitry of the PV cells allow for alongside the current going into the Li-ion cell. You're acting like there's some extra degree of freedom that allows the current and voltage of the solar panels to vary by some method other than ohm's law, but there isn't because the resistance of the circuit is more or less constant. Measure the open-circuit voltage of the PV cell and then the shirt-circuit current, that should give you a rough estimate of the equivalent series resistance (R = V/I), which should be far higher than that of the li-ion cell but probably lower than that of any series resistor you'd add in there to limit current. Subtract the voltage of the li-ion from the open-circuit voltage of the PV cell and divide the lot by the total resistance and you've got an estimate of your charge current.

[Anonymous No.1503774]

>>1503771
>>1503772
Wait, is what's pictured the charge circuit you're talking about? Because that isn't hooked up "directly" at all, those will limit the voltage going into the cell. I wouldn't run those on something that can go as high as 20V since they're fully linearly regulated and will make some heat. Not to mention it might flat-out refuse to work since those solar cells put out such a pitiful amount of current. If there's no linear regulator and the ICs are expecting 5V from the source, then you'll bugger the circuit immediately.
tl;dr use anything other than solar panels for fuck's sake

[Anonymous No.1503779]

Do you think lm317 can do 3 amps for couple of seconds?

[Anonymous No.1503781]

>>1503779
Put a heat sink on it and it can do so indefinitely. Otherwise, no, the heat builds up on the tiny silicon die far too quickly.

[Anonymous No.1503782]

>>1503781
Forever? But it is only rated for 1.5v if heatsink fixed that i am sure the datashit would mention it

[Anonymous No.1503784]

>>1503782
1.5a

[Anonymous No.1503794]

>>1503772
I am trying to measure the AC sine voltage with a probe placed on the top leg of the burden and ground clip placed in the middle of the voltage divider
but i am unable to get a sine, all i am getting on the scope are garbage scrambles

[Anonymous No.1503800]

>>1503794
i meant ground pin placed at arduino GND
expected result should be a positive top half of a sine between 0 - 5V

[Anonymous No.1503804]

>>1503800
Scope input across burden resistor? Where did you get the 100A primary current from?

[Anonymous No.1503805]

>>1503804
it's not across burden, the burden swings between positive and negative, the votlage divider shifts the ground referrence so i measure on the top of the burden and ground of the arduino which swings between 0 - 5V
the actual current is not 100A, that doesn't really matter, for testing purposes i raised the burden resistance to get 5V at peak of the AC going through the CT

[Anonymous No.1503807]

>>1503794
measure current going to burden resistor, measure voltage across burden resistor and then report back

[Anonymous No.1503810]

>>1503689
i try, but that's what half of chips with 4-digit type numbers seem to be these days

>>1503782
typical max output current is rated at 2.2A, but it is also limited by the Vin-Vout differential and heat dissipation

[Anonymous No.1503814]

>>1503782
>Maximum power dissipation is a function of TJ(max), θJA, and TA.
>The maximum allowable power dissipation at any allowable ambient temperature is PD = (TJ(max) – TA) / θJA.
>Operating at the absolute maximum TJ of 150°C can affect reliability.
As stated on the datasheet. More specifically, P = (V_in - V_out)*I_out. The junction-to-ambient resistance is probably about 24 C/W, and for 3A at the minimum 1.25V drop that's 3.75W, or ∆T = 90°C. Add that to ambient, or 25°C and you've got 115°C, which is just below the maximum recommended operating temperature of 125°C even without a heat sink. If you're dropping 10V though you'll need a heat sink with more stopping power.

Though I'll have to admit, if you're asking this question in the first place you should be looking at some sort of switching regulator.

[Anonymous No.1503815]

>>1503810
>>1503814
Oh does output voltage regulation just die upon exceeding ~2.5A? In that case then no dice, perhaps you can get away with a big enough output capacitor.

[Anonymous No.1503818]

>>1503814
you mean buck converter?
i need a constant current so no dice there and the output also has to be super clean, i want to use it to test a laser diode i posted before, before i order a new driver and find out it was the laser that was fried
it is also just for a 2 second test to see if it turns on

[Anonymous No.1503819]

>>1503818
like i literally just need the lm317 not shit itself long enough for my eyes to register any light coming out at all, even .1s at 3A is fine

[Anonymous No.1503821]

>>1503818
Could be buck converter, could be a capacitive switcher (charge pump?).

>>1503819
In that case a large output cap (and maybe a diode before it) should be plenty. Though personally I'd test it at a lower current and/or with a power resistor as your dropper. If you're dropping more than 0.5 times the diode voltage you shouldn't have a problem with thermal runaway.

[Anonymous No.1503822]

>>1503821
i was told that laser diode won't work at a lower current than it's operating current, was that not true?
As in, if you supply 98% of the current the diode normally operates at it won't turn on until you supply between 99% - 100% of its normal current

[Anonymous No.1503824]

>>1503822
I'm no expert at laser diodes, but as an expert in generic laser physics I can tell you that there should be a wide range of pumping power that will result in lasing characteristics. And since semiconductor diodes are pumped pretty efficiently you shouldn't have an issue. At the very least, sending a smaller current to your laser diode shouldn't ruin it if you want to test that.

That advice might be true for the voltage, however. In either case, I'd advise looking for a source other than anons on a jamaican cattle-herding forum.

[Anonymous No.1503826]

>>1503824
Ok i will do it like this:
set up the lm317 with a potentiometer (to regulate the constant current output amount) and then start raising the current from 10ma to 3A slowly and if at any point during the test the diode vomits out any visible photons, i will turn the power off and call it a success

[Anonymous No.1503827]

>>1503826
Put a pushbutton between the 317 and the diode, since surpassing 2A or so for a prolonged time might blow your 317. Good luck!

[Anonymous No.1503831]

>>1503822
If the current is too low, it's like like an LED and is much dimmer than you'd expect from the current. Actual turn-on is pretty abrupt.

[Beppu !n/ZnkB9jUw No.1503836]

>>1503774
>>1503772
The whole thing was the one little solar panel will put out like 0.8V under interior lights, 2.0V with a flashlight on it, and then 5.0V under bright sunlight at the right angle.

So when I put the 4 panels in series, I can get >4.0V under these flourescent lights and then 8V-9V with a flashlight inside and I’m guessing I could get nearly 20V with good sunlight since I got 5V per panel individually. When I had one cell under interior lights (~2.0V) I was getting negative current between the panel and cell. With 4 panels in series, I got 5V-8V under the same lights and actually got positive current to the battery (but low, low current). That was the same with or without that circuit board.

I think I learned that the charging source needs a higher voltage than the battery to charge it, so 2.0V output won’t charge a 3.7V battery. So when I get out into sunlight where one panel would get 5.0V, I’m curious what the 4 in series will do and what that board does to the voltage going to the battery. The board was originally just hooked to one panel that maxed out at 5.0V in bright sun so we’ll see. I was sort of expecting the board to bump up the voltage so even partial sun at 3.0V from the panel would jump up to ~5.0V and charge a 3.7V battery, but it stayed the same. Still learning...

[Anonymous No.1503837]

>>1503836
there are boost converter boards that will step up the solar panel voltage to charge the battery. and even some that'll optimize their current draw to get the maximum power out of the panel without collapsing the voltage. if your setup works then it doesn't really matter though.

[Anonymous No.1503838]

>>1503510
Standing in one place actually ain't healthy (walking is).

[Anonymous No.1503839]

>>1503838
walking wears out joints

[Beppu !n/ZnkB9jUw No.1503840]

>>1503772
>>1503836
See, this board confuses me. Without the little board hooked up, the panels were reading 4.0V. I hook up the panel with no battery or LED attached, and voltage drops to 0.95V at the panel. Orange meter is the voltage at the output to a battery (not connected) which is <0.70V. I expected the board to give a more steady voltage to the battery even with varying voltage coming from the panel, but I guess that isn’t the case.

I need to work on these equations some more to understand it. I’m not sure what that board is doing. It’s the innards of little solar lights so there’s 3 +/- pairs. One for solar panel, one battery, and the third for the LED with a daylight sensor.

[Beppu !n/ZnkB9jUw No.1503841]

>>1503837
That’s what I though, that the output for the battery should be ~5V whether the solar panel is sending it 2.5V or 5.0V. But it doesn’t seem to be the case. Unless possibly the fact that even bright interior lights are a horrible substitute for the sun.

[Anonymous No.1503842]

>>1503839
In moderation it harms you less than sitting or standing in one place.

[Anonymous No.1503843]

>>1503841
>Unless possibly the fact that even bright interior lights are a horrible substitute for the sun.
yes

[Beppu !n/ZnkB9jUw No.1503845]

>>1503843
Maybe the current is just too low. But I’m at 4.1V on the series of panels, which is close to full sun if it were outside with one panel like the original light.

Just tried a second board and it’s the same thing. 4.1V on the panel alone. Hook it up to board, reads 0.95V from the panel and 0.69V from the output to the battery.

And why would the voltage at the panel drop from 4.1V to 0.95V without a battery or light connected?

Sun is coming up so I can go fux wit it outside in a bit I guess.

[Anonymous No.1503851]

>>1503845
the output voltage doesn't increase linearly with available power. pic is an example of a single cell, where a panel will have many cells in series.

>And why would the voltage at the panel drop from 4.1V to 0.95V without a battery or light connected?
the board has its own current consumption independent of the battery. granted i can only see a coil and a PIR on the side of the board in your pic so i have no idea what it's doing underneath.

[Beppu !n/ZnkB9jUw No.1503853]

>>1503843
Welp, it’s darker outside than it is on my workbench, but the sun definitely has some more juice to it. Partly cloudy, dawn, and on the west side of the house so it’s shady and this is what I got.

Solar panels were reading 8V-9V with nothing attached. That still seems like a big drop considering no battery or light hooked up.

[Beppu !n/ZnkB9jUw No.1503857]

>>1503851
Hmm ok I see. I is an retard with electricity. I need some real sun to see what happens but the voltage coming in should be much higher than it originally was and I want to see how that effects the battery charging voltage and rate.

And pic shows the front and back of the boards. A little switch which was turned on a year ago when they were new, daylight sensor in the middle, and then the 3 connections on the other side for the solar panel, battery, and LED. I was assuming the daylight sensor is basically an on-off switch leading only to the LED output and not connected to anything else but I’m not sure.

[Anonymous No.1503858]

So i read the guide on lips and it seems actually pretty easy.
So for one cell:
Don't discharge or charge them faster than 1C, when discharging don't let them fall under 4V, when charging (ideally at 1C) stop charging when they hit 4.2V.
Is that it?
I can't believe so many people are getting blown up, this shit seems easy.

[Anonymous No.1503859]

>>1503858
>lips
*liions/lipos

[Anonymous No.1503861]

>>1503858
>Don't discharge or charge them faster than 1C
usually you can get away with faster but it depends on the cell. check the datasheet and monitor the temperature
>don't let them fall under 4V
kek, no, you'll only get 5% of the capacity if you do this. 3V/cell is usually safe
>ideally at 1C
too fast for most cells. C/4 is almost always safe but check the datasheet and monitor the temperature
>can't believe
I habeeb it

[Beppu !n/ZnkB9jUw No.1503862]

>>1503858
>don’t discharge under 4V
Huh? That one seems too high. I thought it was cool to go 3.0V- 3.5V.

There is some stuff about storage too. I have heard you want to store them at about 70% charged and there was something else about not letting them sit for more than a couple months with a little cycle/charge.

[Anonymous No.1503864]

>>1503857
more reverse engineering always helps but i don't see a switcher on that board since what looked like a coil is only a switch. so i guess that board is just a linear charge controller of some sort. if you care about efficiency you could try to find a switching charger board, ideally with MPPT. unfortunately googling that will be hard since the results will be full of 500000mAh solar packs and control units for home solar.

[Anonymous No.1503866]

>>1503862
>>1503861
that was a typo i meant 3v

one thing that sucks is that i will have to add some sort of charge control into all my projects to make sure they don't drain the cells under 3v since i don't want another Chernobyl on my hands

[Anonymous No.1503870]

>>1503861
>I habeeb it
Damn, that is some vintage meme there, i actually had to look it up.

[Beppu !n/ZnkB9jUw No.1503872]

>>1503866
I have been ripping apart these $5 Walmart power banks for that purpose. It’s just a 3350mAh 20650 battery and that little board with a USB out and Micro USB in. Should give you charging protection if you charge with the micro USB, and discharge protection going out through the other USB (or soldering to those points like I did).

Only issue I had was the boards not exactly liking my blinding bright LED projects, like they’re pulling too much current for those 0.5A chargers. I grabbed a larger 1.0A charger but have yet to rip that one apart.

I’m sure there’s a better way to do it, but this was easy.

[Anonymous No.1503876]

>>1503866
>>1503872

>have to add some sort of charge/discharge controller to my projects

Here you go. cost about .50usd

[Anonymous No.1503890]

>>1503866
there are protection boards that will help keep you from setting the building on fire, but they're not designed to protect the battery itself from such damage that you might safely recharge it. fortunately, 3-pin voltage supervisors are available, and can be combined with a p-MOSFET and a large resistor to shut down a load early enough to allow safe recharging, without taking up a whole lot of space on your board
protection-only boards are like 5 for a burger. better to apply them nearest the battery in case of accidental shorting or the like. also there's no charge circuitry included so you might have to take the batteries out and put them into an external charger
protection boards with a built-in TP4056 micro-USB linear charger are easily gotten and pretty cheap https://www.aliexpress.com/item/TP4056-5V-TENSTAR-ROBOT-1A-Micro-USBenis-18650-Lithium-Battery-Charging-Benis-Protection/32818008660.html

[Anonymous No.1503891]

>>1503890
>benis
every time

[Anonymous No.1503897]

>>1503890
charging is fine, i will never charger them inside of my devices, i only need a discharge protection so they don't dip below 3V or draw too much current, if that board you linked can do that which as i look at it it can that is exactly what i needed, i am going to order a pile of them now

[Anonymous No.1503914]

>>1503897
>dip below 3V
those only cut off at 2.5V or less. when much below 3V, it's unsafe to charge them without special precautions and procedures if at all, which is why I suggested using a voltage monitor and a MOSFET to protect your battery to easily charge again

[Anonymous No.1503936]

what are the downsides of using large voltage divider resistors?
Surely there must be some or everyone would be using like 100Mohms

[Anonymous No.1503937]

>>1503936
whatever ADC or op amp is reading it will have a bias current. that'll pull the divider's voltage up or down, and it's worse the larger the resistors are. also resistor thermal noise increases with value but that doesn't matter as much.

[Anonymous No.1503938]

>>1503937
I see. So about 20k should be good for a battery powered arduino project?

[Anonymous No.1503939]

>>1503938
20k is perfect. iirc the arduino's atmega datasheet calls for a max of 10k input impedance on the ADC, and two 20k resistors in parallel equals 10k.

[Anonymous No.1503943]

>>1503936
lookup "thermal noise", the effects of thermal noise are worse as the resistance increases.

[Beppu !n/ZnkB9jUw No.1503950]

>>1503876
>>1503890
>>1503891
As I mentioned, I’m sure there is a better way to do it, but I was trying to play with the thing this weekend, not April 7 - May 22, 2019 if that China shipping goes smooth.

They actually worked suprisingly well once I worked through the overcurrent bullshit and dumped about 8 hot glue sticks into the thing.

[Anonymous No.1503958]

This looks really cool, do you think it's worth making?
https://www.thingiverse.com/thing:1467855
it is for leg bending

[Anonymous No.1503959]

>>1503958
i wish i had one of these. if you know cad though it might be better to design a two part die.

[Anonymous No.1503965]

>>1503959
You mean cut it in half? I can do that easily in blender. But that seems like you now have to take care of two items instead of one

[Anonymous No.1503967]

>>1503965
i was thinking of having a top half to the die so your big meaty claws don't get in the way of forming anything 0.3" spaced or smaller like DIPs or 1/4W resistors. i haven't used one of these before though so maybe it's not a problem.

[Beppu !n/ZnkB9jUw No.1503998]

I was really hoping to see more sparks by stripping the wires and throwing all of these in a pile but I guess they’re too dead.

Maybe they need some gasoline

[Anonymous No.1504003]

I often noticed that when i fucked up and something starts smoking, then if i turn the power off quickly, and fix the mistake, the thing than smoked works just fine. Do most components not mind having a bit of a smoke break if you don't over do it?

[Anonymous No.1504004]

>>1504003
resistors can definitely handle some abuse. i've never seen a semiconductor survive after the smoke escapes though.

[Anonymous No.1504006]

>>1504004
Yeah, the only option after that is to take it to a repair shop and have them pump the replacement smoke in and seal the tear so it can't escape back out.

[Anonymous No.1504011]

>>1503959
It's breddy cool, smaller than it looked. Just buy a printer man, they are cheap as fuck, i am sure my friend ping pong can get you one for under 500 burger coins.

[Anonymous No.1504036]

>>1503998
stick a nail through it.

[Anonymous No.1504046]

>>1504011
Bending gauges are dirt cheap, less than one € coin.

[Anonymous No.1504063]

>>1503950
I sometimes get shipping in 10 days, and I'm in fucking New Zealand. I think that's through Yanwen Air Mail. Not to mention shipping times are typically within 1 month for small PCBs and such, and you've been doing this for at least ~2 months now.

So did you measure the short-circuit current under fluorescent lighting?

[Anonymous No.1504111]

>>1504046
seems really easy to make with some thick cardboard or some random trash

[Anonymous No.1504113]

>>1504111
Well, it is just as simple as it looks like.
That said, if you're such a jew that 1€ is too much, then there's no pressing need to get that thing.

[Anonymous No.1504119]

I`ve got a bunch of motors and steppers from hard drives and readers. Most things I see on the web about reusing them involves buying a cheap arduino driver shiel or something similar. Why people don`t make their drives? It isn`t viable? I`m pretty sure I can make two low power H-bridges to control the steppers, maybe I`m missing something.

[Anonymous No.1504151]

>>1504119
you can make manual bridges or you can buy ICs for it. people buy shields because they don't require soldering or thought.

[Beppu !n/ZnkB9jUw No.1504165]

>>1504063
>So did you measure the short-circuit current under fluorescent lighting?
Nah, I was busy shooting those used Li-Ion cells with my BB gun.

3 of the 4 were swollen but still had a low voltage reading and those didn’t do shit. The 4th gave like 0.006V and wasn’t swollen and that one let out a nice puff of smoke when I sent a copper BB into it.

[Anonymous No.1504166]

>>1504165
>0.006V
looks like usable energy to me

[Anonymous No.1504169]

Trying to design babby's first switching benchtop power supply, of the non-isolated buck topology, preferably with a current and voltage knob that works just like those LM2596 modules. I'm trying to design it discretely from op-amps and comparators and the like but I can't seem to get the variable duty-cycle sorted at all. Do I need an instrumentation amplifier? I'm guessing rail-to-rails help, but is there an easy way to not require them? In general, I'm asking for pointers to circuits or sites. Perhaps looking at a buck converter IC's datasheet for the equivalent circuit would give me something usable, but I suspect they'll be a little complex. Pic related.

[Beppu !n/ZnkB9jUw No.1504171]

>>1504166
Those lights were outside for like a year and a half. I think the one reading 0.006V had a corroded terminal and was the only cell that was actually useful if the terminal were fixed since the other three were swollen.

Two of the swollen ones were like 300mA and the third one was 1.7V so I was hoping the higher one would do something cool when I shorted and shot them but nothing happened to those so I tossed em in the recycling bin.

[Anonymous No.1504193]

>>1504169
Is there a site or a book that can help me read schematics like this? I didn't see any info in OP, unless its in the books that are recommended.

[Anonymous No.1504215]

>>1504193
It's all fairly representative, aside from the comparators and op-amps and transistors and linear regulators. Know how those components work, alongside simple digital logic elements such as gates and latches and flip-flops, and also circuit elements like capacitors and such, and you've got basically all you need to know. The boxes "driver", "thermal limit", "current source bias", etc. are representative of what are probably fairly common building block circuits, at least in the context of the circuit.

While "driver" doesn't mean much on its own, seeing that it's connected to a latch and thermal limit and feeding some power transistors in a switching converter circuit is enough to know that it's a circuit made to turn transistors on and off as efficiently as possible according to the latch output, and has a thermal override input that will turn off the circuit if the temperature gets too high. From this I'd simply use an AND gate to combine the latch and thermal limit inputs and feed that into a BJT totem pole.

[Anonymous No.1504225]

>>1504215
Most of that sounded gibberish to me. When I see the shematics the R1, R2 etc.. I assume those are resistors and their values. I can identify SOME stuff, but its merely guess work for me. Im sure I will get better at it over time.

[Anonymous No.1504253]

0dBFS is the maximum possible level in digital audio but what voltage does that translate to?

[Anonymous No.1504264]

>>1504253
90KV

[Anonymous No.1504282]

>>1504225
>Most of that sounded gibberish to me
In that case, learn the basic building blocks and work your way up from there. I took physics uni papers on basic (analogue) electronics and digital logic, but just looking at and making simple circuits, googling what you don't understand, and advancing to a more complicated circuit. Big Clive is a great channel to see what fairly simple circuits are out there commercially, and EEVblog is pretty good for more complicated circuits and a bunch of other drongo.

I'd start with ohms law and kirchoff's laws, move on to diodes and BJTs, perhaps touch on some transistor amplifier topologies, then probably go to op-amps. One of the books in the OP would be pretty good for this. LTspice is free and a powerful tool for messing about with circuits once you understand its quirks. On the other hand, Logisim is pretty neat for digital logic.

[Anonymous No.1504285]

>>1504253
There's no single standard, it varies depending on where you are and what you're doing.

[Anonymous No.1504291]

>>1504285
Well it has to translate to something. If I take a 0dBFS peak waveform and put it right through a DAC and terminate it into a high impedance load like a buffer amplifier with no other analog voltage amplification what level should I expect to get out?

I don't understand how this can vary.

[Anonymous No.1504293]

>>1504291
Depends on the DAC, read the datasheet.

[Anonymous No.1504339]

Can you recommend me some good ADC module for arduino?
Basically i need something that has its own precise voltage reference and is able to get me good precise 12bit adc readings with voltages around 0 - 5V

[Anonymous No.1504350]

>>1504339
Doesn't have to be super duper precise, just good enough to allow me to sample output from a CT and get RMS

[Anonymous No.1504363]

>>1504339
ADS1115/ADS1015? They're 16 bit and come on little PCBs for those unable of thinking for themselves. But why 12 bit?

[Anonymous No.1504366]

>>1504363
12 bit is minimum i need, more is of course fine
10 bit on arduino is not enough
Also how does it handle the voltage reference? on arduino using vcc 5V reference blows up the second you try to run it on batteries, and the internal 1v1 ref is garbage since it is only 90% precise

[Anonymous No.1504367]

>>1503891
yes. every time

>>1504169
>variable duty-cycle sorted at all
the 150kHz oscillator generates a sawtooth wave. the amp marked "GM" (transconductance, which is a bit fiddly but all you need to worry about is that the gain is relatively low) amplifies the difference between the reference and the feedback, which changes the duty cycle of the square wave out of the comparator

>>1504193
oh and that's a block diagram

>>1504291
FS, which is chosen by the designer

>>1504366
>only 90% precise
so? calibrate it out in software. or with a trimpot
>Also how does it handle the voltage reference
RULE 1. read the fucking datasheet

[Anonymous No.1504374]

>>1504151
They are ok or it`s stupid to make them instead of buying a IC?

[Anonymous No.1504384]

>>1504374
imo it's a waste of time to create discrete drivers for motors unless you need more power than an ic can give. there are a few well-established h-bridge chips with logic inputs that will save you the tedium of wiring up a dozen discrete components. even for unipolar steppers, the uln200x is an array of transistors with integrated base resistors all in one package.

for hard drive spindle motors (BLDCs) you'll need supporting circuitry to go with your driver anyway though since unfortunately there's no dedicated ICs available in through hole packages.

[Anonymous No.1504498]

I would like to measure the internal arduino voltage reference which is 1.1V to see how much it really is but i am unsure of how to do it.

I don't think i can access the reference directly with a multimeter so, what if i take some voltage 1v or less that is stable, measure exactly how much is it with a multimeter, then connect it to arduino analog pin, and print out the analog value.
So if i know i connected 0.55V to the pin and i get back a reading of say 490, when i know i should get 511 (half of 1023, since i am suppling half of voltage) then i can use the 490 to calculate the real internal reference voltage.
Would this be a good method to do it?

[Anonymous No.1504502]

>>1503555
>them timer chip digits
I like the working height of a standing bench. But I prefer them with at least of portion of it to be open underneath for when I want to sit at the bench on a stool and rest my feet/legs underneath on a foot rest or a wide-open bottom shelf
other wise you are forced to stand always, or you have to sit side-saddle in front of the bench with no place to put your feet under there

[Anonymous No.1504508]

>>1504498
>I don't think i can access the reference directly with a multimeter

is this the internal ADC voltage ref? my pdf for ATmega8535 says pic related -> you can measure it at AREF. I tried to get a pdf for a newer AVR but google was not nice.

[Anonymous No.1504510]

>>1504508
So i set the analog reference to internal and then i measure the voltage between ground pin and the aref pin? that would be so much easier

[Anonymous No.1504511]

>>1504510

that's the way I interpret it.

fucking microchip won't show me the 328p data sheet.

[Anonymous No.1504512]

>>1504511
well there is easy way to find out, if the aref pin will measure around 1v1 instead of 5v then it means it's working

[Anonymous No.1504514]

>>1504512

yes.

finally found one with 1.1 ref and it has the same wording, so it should work.

[Anonymous No.1504542]

>>1504367
>FS, which is chosen by the designer

Well that's annoying. Any idea what the typical full scale voltage of a DAC that outputs line level PC audio would be?

[Anonymous No.1504549]

>>1504542
line level standard is 2Vrms for consumer gear. but any given DAC might not be able to reach those peaks

[Anonymous No.1504564]

can someone recommend me jelly bean parts that are easily found? People that work in local shops don`t know anything, they just give you part name you ask for.
>CMOS arrays for use in a h bridge 0.5-1A
>eeprom serial memory chip >1kb

[Anonymous No.1504566]

>>1504514
So i tried it and uploaded a sketch with just analogReference(INTERNAL), but the voltmeter shows 2.5mV so it doesn't seem to be working

[Anonymous No.1504581]

>>1504566
Post contains no useful information.

[Anonymous No.1504608]

>>1504564
>CMOS arrays
if you want some decent jellybean switching MOSFETs, consider the AO3400 and AO3401. very low Rds(on) and Vgs (0.085 ohms at 2.5Vgs), but be prepared to deal with SOT-23 packages
if you want a complementary pair in one SO-8 package, try IRF7309 IRF7343 Si4532 Si4542 AO4600
if you want parts that are handy to breadboard, you can't beat IRFD024 and P-ch IRFD9024, if you have enough gate voltage to get them open. otherwise the good old IRF520 and complement IRF9520 ain't bad to have around, even though kinda large
>eeprom serial
24c16 is a 2kx8 I2C device, 24c32 is 4kx8, etc. on up to 24c512 25c16 and M95160 are 2kx8 SPI devices and so on, and there are 25q devices whose type numbers reflect Mb instead of kb

>>1504566
did you turn off digital output on that pin? did you power up the reference (if needed)?

[Anonymous No.1504610]

>comp sci student
>bought a stm32 dev board, motors, md, and other components
>the code is easy
>can't figure out how to design a circuit
feeling like a brainlet tonight

[Anonymous No.1504611]

>>1504608
Thanks friend. Don`t you know any THT arrays? I have a bunch of irfz44n and the complementary pchannel guys, but they seem like an overkill for what I`m doing. I also have some 20$ IGBTs a lab threw away (600 V, 75A kek). I`ve found some 24C16 chips already, gonna buy them in my next shipment.

I`m making a lab supply until my scope arrives, then I`ll make something to make signals, then a laser engraver. Feels like minecraft..

[Anonymous No.1504625]

>>1504611
>THT arrays
I don't know of any. I doubt there are very many per se. the closest I've seen are TPIC6B273 open-drain 100mA drivers with latches built in. I wish there were wider, more powerful, plain old MOSFET arrays even in SOIC
usually I just build my own when I need a DIP. you could get some of those handy-dandy adapter boards and put your SO-8s on them along with some machine pins, then treat them as PDIP-8s for all practical purposes. it might even be cheaper than buying PDIPs proper, in some cases
HVMDIP (4-pin 0.300" DIP) are stackable, look for some part numbers in that package using the parametric selector at http://www.vishay.com/mosfets/

[Anonymous No.1504630]

>>1504498
set the supply rail as your reference voltage. measure aref with that. then measure your supply voltage and do the math to figure out aref.

[Anonymous No.1504632]

>>1504625
>>1504625
What does it mean to be stackable? Literaly one on top of the other? I don`t see how that could work.

[Anonymous No.1504639]

>>1504630
iiuc, the internal referenced can't be switched onto an input, but it can be switched out to an external pin for decoupling if it is set as the ADC's reference

>>1504632
side by side in a 0.300" DIP footprint

[Bepis !n/ZnkB9jUw No.1504655]

>>1504639
These are the things you call breadboards? So what’s on the back? I’m assuming lots of alligator clips while you build and test, but then do you wire and solder or what? Or this isn’t permanent so no solder but you test on here before having the PCB made?

[Bepis !n/ZnkB9jUw No.1504658]

>>1504610
Kek, my little brother can code and has a degree but I gave him a multimeter the other day and he was like “So what do I do with it?

[Anonymous No.1504661]

>>1504639
thanks
>>1504655
>>1504658

1.2/10, please remove your trip and die

[Anonymous No.1504667]

>>1504658
it took me about an hour to kinda grasp the difference between volts, current, and resistance.
Still not sure I actually understand it.
>don't know why I thought hobby electronics would be somewhat easy

[Anonymous No.1504675]

>>1504632

>>1504639
>side by side in a 0.300" DIP footprint
no, no, no

[Bepis !n/ZnkB9jUw No.1504733]

>>1504661
>12/10
Thanks bro

>>1504667
It took me a couple fuses to realize you don’t measure current on the DMM by touching red to red and black to black. And about 10min before I realized my desk fan isn’t running at 900W and X10 isn’t a model number.

[Anonymous No.1504744]

>>1504367
>the gain is relatively low
So this is the problem. I thought just using an op-amp wouldn't work too well. Using an instrumentation amplifier should work, right? Or a differential amplifier if the signal is buffered enough. I'll see about combining the current and voltage feedback lines once I've gotten each sorted individually, thanks.

>>1504655
If you don't at least know what a solderless breadboard looks like, least of all own one, you don't belong on this thread.
>inb4 you buy one for more than $8

[Bepis !n/ZnkB9jUw No.1504747]

>>1504744
>if you can’t run before you can walk, you don’t belong on MY 4CHAN!
Bad news bro, this is 4channel now.

[Anonymous No.1504765]

>>1504744
Tripfags don't belong on this site let alone in this thread.

[Anonymous No.1504782]

>>1504744

[Anonymous No.1504783]

>>1504747
Says the retarded cunt who can't into a google search. But keep embarrassing yourself faggot. You're doing great things for your brand.

[Anonymous No.1504794]

*reapplies filter*

[Anonymous No.1504838]

What is the easiest way to precisely sense mains AC voltage with arduino?

[Anonymous No.1504841]

>>1504838
voltage divider thing > adc
pic related looks pretty good

If you just want to sense whether it's live instead of measuring its amplitude, then either rectifying it and dividing it down and/or some form of capacitive sensing to isolate the ardy should do the trick. Watch that you don't get any nasty ground loops, but that your arduino's ground is properly referenced relative to the mains. I'd recommend using the earth lug for this if you have the luxury (i.e. aren't pushing any current through it), but otherwise the neutral pin should work fine.

[Anonymous No.1504846]

>>1504841
uhh.. connect arduino directly to mains ac voltage with nothing between but a couple of resistors? that is scary as fuck

[Anonymous No.1504853]

>>1504846
It is scary. Would not recommend unless you can have the arduino well away from anything you can touch including anything connected to the arduino. Also make sure voltage rating of R2 >> the actual voltage.

Going via a transformer can be much safer but adds distortion and phase change if you care about that.

[Anonymous No.1504860]

>>1504853
Isn't phase change only a problem if you measure multiple phases?
What i need to do is measure peak sine and bottom bitch sine, so i can calculate RMS voltage
i will definitely want to use a sense transformer, but i am not sure which one, all i can find are current sensing transformers

[Anonymous No.1504865]

>>1504860
unless where you live is pretty fucked mains has symmetry, you can just get the positive peak.
You could use
>resistive voltage divider
>some fuckery with opto isolators
>transformer
>capacitive power supply thing

[Anonymous No.1504867]

>>1504865
I was toying with the idea of measuring the mains voltage with a multimeter and just hardcoding that value. I mean, i wouldn't think the mains voltage in developed worlds wobbles too much.

[Anonymous No.1504869]

>>1504867
It does change a little, what is usually very controlled is the frequency. You`d have to look up what are the regulations you have there and see how it is allowed to vary. But it`s not more than 5%, which is the same you`ll get with your resistors so meh

[Anonymous No.1504873]

Excuse me, i only use the highest quality chinesium 5 band resistors with 1% tolerance.
I just want to measure current for my appartment so it should be good enough. Basically i donẗ want to be off by more than two kilowatths per month,

[Anonymous No.1504875]

>>1504873
>>1504873
then you have to measure V and I and multiply each one point by point to discover you`ll be paying less than you should because of all the harmonics you cause with electronics that aren`t measured in the analog meters.

[Anonymous No.1504876]

>>1504875
Also there are indirect current measuring methods for mains. You just pass the wire inside the sensor and call it a day.

[Anonymous No.1504895]

>>1504639
>iiuc, the internal referenced can't be switched onto an input, but it can be switched out to an external pin for decoupling if it is set as the ADC's reference

this is correct; it appears on the AREF pin, which is labeled REF on my nano.

>>1504566
>So i tried it and uploaded a sketch with just analogReference(INTERNAL), but the voltmeter shows 2.5mV so it doesn't seem to be working

per pic-related, using assembly language on a nano (ATmega328P), if I set refs0 and refs1 for AVcc, voltmeter says 4.82 volts (Vcc reads 5.04). if I set them for internal 1.1volts I get 1.17 volts, so it works as it should.

Note that the register that controls this is the ADMUX register which also selects the ADC pin to read, which means that the Arduino code will write to this register when you call analogRead(). It's possible that the Arduino code screws this up, and I don't see any way to use the registers directly unless you go to assembly code.

Note that if you are using an Arduino Mega you use INTERNAL1V1 instead of INTERNAL for analogReference().

[Anonymous No.1504897]

>>1504869
1% resistors are cheap as shit. Can also calibrate value in software.

[Anonymous No.1504898]

>>1504675
end-stackable, that is

>>1504744
you don't need that much buffering, really, just high enough resistances on your diff amp. a full inst amp is overkill for accuracy and probably underkill for gain
you could get a proper transconductance amp, but they're not exactly jellybean parts
>combining current and voltage feedback
an error amp and comparator for current, in parallel with those for voltage, wired such that either one can turn off the output transistor

>>1504873
>two kilowatths per month
that's less than 3 watts. sure you're ready to tackle that kind of precision?
also what
>>1504875
said about power factor without actually saying power factor

[Anonymous No.1504899]

>>1504898
power factor is linear, analog readers don`t get that, but harmonic distortion caused by switching isn`t linear and he will also get that. and depending on how he lives that can be significant.

[Anonymous No.1504901]

>>1504899
Measuring RMS current * assumed constant voltage can be would be a reasonable guess, should read above your real power.

[Anonymous No.1504912]

>>1504875
>>1504898
I think i can get pretty precise actually.

I will use a good proven CT with a high as possible 6band precise burden resistor so that the voltage is as close to the maximum arduino can handle as possible at maximum current my apartment draws, to get maximum resolution.
I will use an external ADC with 16 bit resolution instead of the shitty 10bit arduino one and use an external voltage reference IC with 0.5% precision for the ADC to compare the measured voltage to and with all this i am pretty sure i should be able to easily fit within the 3kwh
and if not, then then electronics as a whole are just a joke and nothing more than nature's random value generator

[Bepis !n/ZnkB9jUw No.1504921]

>>1504794
Why did you ever remove it?

[Bepis !n/ZnkB9jUw No.1504927]

>>1504783
>>1504765
Can you pls explain it to me? I need you to make airplane noises as you’re doing it.

[Anonymous No.1504938]

>>1504927
No really, fuck off shit dick. You're not funny, you're not ironic, and now you're shitting up our thread as well.

[Anonymous No.1504945]

>>1504912
Do you have physical access to the power meter for your apartment? If so, a very accurate approach would be just stick a light sensor on the meter's pulse LED and count the flashes.

[Anonymous No.1504957]

>>1504945
>accurate
in the sense that it will agree precisely with what the power company bills. if that's what electric meter anon is interested in, maybe that's the better way to go

[Anonymous No.1504959]

>>1504945
he is lookin tomeasure the peak, not frequency or zero crossings.

[Anonymous No.1504961]

>>1504945
couldn't do that anyway, the meter is locked away in a cabinet owned by the power company
they wouldn't let me putting anything there anyway because they do not want you to know how much energy you are using for obvious reasons

[Anonymous No.1504965]

>>1504961
>hey do not want you to know how much energy you are using for obvious reasons

doesn't it show that on the bill

[Anonymous No.1504975]

Anyone got i2c to work with ESP32?

[Anonymous No.1504981]

I have a floor lamp with a dimmer that's acting weird. First it started pulsing erratically at a very low frequency (with periods of normal function), and now it will immediately go to fully power as soon as I turn the knob. Any ideas what the issue may be?

[Anonymous No.1504984]

>>1504965
the bill is annual, it's just a big mistery lump number, they don't want you to know it in real time so you go like "wow that second fridge sure eats loads of power, better turn it off during work days"

[Anonymous No.1504989]

>>1504984
>the bill is annual,

are we talking about residential power or something else. I have never heard of annual power bills in the usa.

[Anonymous No.1505008]

>>1504981
triac or diac = ded

[Anonymous No.1505010]

>>1504989
>usa
I'm not fat. Electricity bills here are annual for pretty much everyone.

[Anonymous No.1505011]

>>1504975
Yes.

[Bepis !n/ZnkB9jUw No.1505014]

>>1504984
>>1505010
Is electricity there a private company or gov’t owned?

[Anonymous No.1505026]

>>1505014
Both

[Anonymous No.1505050]

>>1505026
>Both

so where exactly do you live, where the power company is both a private company and a public utility, and they charge once a year.

that annual billing would go over great in america, where losers would rack up a thousand dollar bill and get evicted because they don't know how to save money.

[Anonymous No.1505052]

>>1505050
there is a big government owned energy company and also private subcontractors

[Anonymous No.1505053]

>>1503466
What's with all the cabinets? Are there really that many different varieties of small parts?

[Anonymous No.1505054]

>>1505053
You need 50 different drawers just for resistors.
And then you have diodes, caps, chips and other chips every one needing 20 or more drawers

[Anonymous No.1505055]

>>1505053
Haha. Go check the component distributors mentioned in OP and see how many different resistors they have. Or transistors or whatever.

[Anonymous No.1505056]

>>1505054

>he doesn't have a little plastic box with 50x each of 100 values smd all sizes in little zip bags

[Anonymous No.1505124]

>>1505053
yes. ICs alone could take up half of that space plus the undershelves

>>1505056
>not keeping your SMD components on the cut tape

[Anonymous No.1505177]

What on gods earth are VR3:VR0? this is a pic675, I`m trying to understand teh comparator

[Anonymous No.1505184]

>>1505177
>VR3:VR0
a 4-bit field from some config register

[Anonymous No.1505196]

>>1504898
>an error amp and comparator for current, in parallel with those for voltage, wired such that either one can turn off the output transistor
Thanks for this tip!

>>1504957
>>1504959
If it's an old mechanical meter, those do measure actual power instead of apparent power, so would be perfect to slap an IR LED + phototransistor to.

If you're trying to accurately measure the real power, power factor, etc. then a current transformer and sense transformer might not cut the mustard due to phase shifting. There are safety features you can add to the purely resistive method >>1504841 that shouldn't be a problem. Namely a 4.7V zener from the ADC to 0V, and a PTC/resettable fuse before R2. A MOV between the PTC and R2 going to ground will prevent any big voltage spikes from lightning, and adding something like a choke or capacitor to get rid of the normal-mode high-frequency components of lightning that might pass the MOV might help also.

Put a current shunt resistor and an op-amp to get its voltage up to a reasonable level on the neutral line, and you shouldn't have any troubles with high voltage.

[Anonymous No.1505276]

>>1505011
Post code

[Anonymous No.1505333]

>tfw need to do shit on arduino
>there is a library widely used to do shit
>open it and copy the code i need
>the code looks like a clusterfuck with mismatching types, ints declared as doubles, doubles as flotals, some longs as doubles and other longs correctly as longs, adc is calculated with as a 1024 top value, even though it's zerobased so it goes up to 1023
the problem is that i don't know if this code is complete shit that will greatly benefit from me fixing and optimizing ir... or the code is next level genius which uses quirks of the platform to stretch the way the types and shit work with each other and stretches them to the absolute limit by intentionally mismatching shit and therefore fixing it will actually make it worse
i mean it's a hugely used library... it's can't be fucking shit

[Anonymous No.1505337]

>>1505333
If it's hugely used, then there's a chance that the guy who originally made it knew his stuff, but the guys who improved it later were retards.

[Anonymous No.1505340]

>>1505333
It's arduino, go in with literally zero expectations, and come out with even lower ones.

[Anonymous No.1505341]

>>1505337
Opinions are a dime a dozen.
Knowledge means work.
Belief is for free.

[Anonymous No.1505342]

>>1505337
Well it's mainly used by normies who don't know shit about electronics and are just following a tutorial to make a cheaper alternative to an expensive commercial solution.
I guess i will try to optimize it, since i just don't see how that crappy code could be covert high iq shit.

[Anonymous No.1505376]

Sup /ohm/, think I just made a pretty gutting discovery.

Pic related is the mainboard of a Commodore 8032, deader than fuck. I have been trying to establish the power rails at least are present but this is the only schematic I can find for the board and there are some different part numbers on the schemo vs the board.

1. Does anyone know of a good lookup site for substitute regs? I remember back in the day doing electronics in high school we had thick arse books full of substitution tables. All I am finding in google is guides on what parameters you need to cross check, breakdown voltages, Pd, Ic etc. as opposed to a "plug in part number, receive substitute p/n" kind of gig. I want to make sure the 7805's are equivalent to the 304-5's and aren't a bad hack by the previous owner (unit came from a ham shack at a deceased estate and shows all sorts of signs of customisation, including a new row of keys above the factory one for RTTY functions)

2. I bypassed the rectifiers and am injecting 9V @ 30mA while I test, enough to run the regs but not enough to let smoke out of the chipset if shit is wobbly. I am however seeing 9VDC on the output of these regulators, which constitutes the gutting discovery. If that is the case all components on the board have likely copped 22VDC/2A unregulated to their VCC from the factory power supply, which is of course the equivalent of a nuke from orbit. All three are exhibiting this behavior however. I would consider it highly unlikely that all three could fail in the same manner unless they somehow break down internally after 20 years of doing fuck all? I seem to recall as I was learning about electronics in the late 90's that TO-3 package regulators could catastrophically shit the bed in this manner and were to be avoided like the plague for TO-220 packages if possible when designing/building, can anyone verify any this?

[Anonymous No.1505396]

You need to prevent against >5.05V overvoltage but you only have 5.1V zeners, wat do?

[Anonymous No.1505398]

>>1505396
Use something else

[Anonymous No.1505408]

>>1505376
http://www.6502.org/users/andre/petindex/boards.html#2

[Anonymous No.1505410]

>>1505376
>1. Does anyone know of a good lookup site for substitute regs?
The term for that would be "cross reference", but they kinda went out of fashion at some point.
LM340 and 7805 aren't strictly identical, but they're close enough that TI has a common page for both of them. Also, considering the data codes of your regs says 1981, they're quite likely original.
>I am however seeing 9VDC on the output of these regulators
I'd double-check your ground connection first.

[Anonymous No.1505441]

>>1503466
Is this connection to blame for really slow internet?

[Anonymous No.1505457]

>>1505441
No? What kind of internet (DSL I presume) and speed are you paying for? And are you see those speeds?

[Anonymous No.1505461]

>>1505457
AT&T Uverse
>Speeds up to 100Mbps.
>Connect up to 10 devices at a time.

It's usually been ok to stream on 2 devices, but it's always hit or miss.
I remember the tech hooking it up a few years ago and saying something about a depreciated line in the garage, but I can't remember what it was. Thought I'd start at the entry point and work my way downstream,

[Anonymous No.1505467]

>>1505461
Another.
Wife is streaming Netflix, but that's all that's going on in the house and these speeds are pretty common (wireless, 10ft away from the modem).
In the pic above, only the green and white are live. The white to the right is the old telephone that was snipped.
From this box, the Cat-58 runs about 12 feet through the garage and into the modem.

[Anonymous No.1505473]

>>1505461
>>1505467
Cricket 4G is faster LOL

[Anonymous No.1505483]

>>1505467
Cat-58? I've never heard of that before.

[Anonymous No.1505494]

>8 pin pic
>4 channel ADC
>want to control 4x7 segment displays
>even with multiplexing the ADC inputs and the BCD outputs AND the displays I`ll only control 2
reeeeeeeeeeeeeeeeeeeeeeeeeeeee, why I`m autist and bought this shit instead of an arduino, I`ll have to use two pics ree

[Anonymous No.1505497]

>>1505483
You're prolly busting my balls b/c it should be a letter, but that looks much more than an 8 than a B or E.
24awg, outdoor rated.

[Anonymous No.1505498]

>>1505497
Yeah, that's an E. As for the "depreciated cable" where did the tech say it was? MSPaint me a map if you need to.

[Anonymous No.1505499]

>>1505497
That's Cat5E

RJ45 wires are usually either Cat5 "letter" or Cat6, the coating is stronger in cat6 making it harder to bend the wire.

[Anonymous No.1505500]

Okay so i got my electricity metering working, i am now accurately sampling watts every second.
But how do the official meters store this?
How do they take watts and turn them into watt hours?
Do they use some sort of averaging?
Should i just have one number and add watt second readings to it?

[Anonymous No.1505501]

>>1505498
>>1505499
Ahh, thanks. I know y'all aren't customer service and I appreciate you taking the time to help.

Cat 5E goes from the exterior box originally pictured, through the brick, around the garage, and into the living room plate which is where the Ethernet cable attaches to the modem.
Now I have no clue what the hell the tech was saying 5 years ago - maybe trying to sell me a BS upgrade. I guess I'll check the connection at the plate as well as the hardware.

[Anonymous No.1505502]

>>1505501
You're welcome, but what the hell is a living room plate? Honestly, that sounds like a potential problem in itself.

[Anonymous No.1505503]

>>1505441
Open up the door to that block (orange lever on the left)
Take green+whitegreen out of the TIP side of the connector (the row of orange levers).

Snip off just the tip, like .5 inches of the green/white gr. Because its mangled by the orange connector.

Actually dont use th ast orange one, go to the next available one.

Put the white on the left hand of the small orange lever, put tge green in the right. There's little grooves in the bottom for the pair

Push the orange lever down. It will resist slightly as it bites into the pair. Tug gently on each wire, they should be secure.

Explanation: your tech was retarded and connected the + and - of the green pair to the - of the incoming dsl signal.

This shorted your line and badly attenuated your dsl but because it's such high frequency you could actually still sync up across the short.

[Anonymous No.1505505]

Am >>1505502
This guy is probably the most right >>1505503

More likely than not, retarded lineman being a tard mangled installation. Shit happens. Easy fix as above.

[Anonymous No.1505506]

>>1505461
The deprecated garage line is the red green yellow black in the back that's not connected to anything.

Its slightly strange he used the green pair to run your service, maybe open your phone jack and make sure it's on green. Otherwise its outside your house.

[Anonymous No.1505513]

Been using pic related's 0-15VDC power supply.

May have overloaded the fucker a time or two. Now it only outputs 28-29V and the voltage adjustment knob no longer works.

Any clue how I fucked it up or how it can be restored?

[Anonymous No.1505518]

>>1505494
you might still have enough pins left for an I2C bus on which you can place up to eight PCF8574 ICs to drive your display
>why I'm autist
or you could use two PICs, which is hardly a terrible thing

>>1505500
>How do they take watts and turn them into watt hours
sample at regular intervals, add samples together, and divide by how many samples you take in an hour. so, averaging, of a sort

>>1505513
the pass transistor has probably failed short

[Anonymous No.1505521]

>>1505518
I need to measure two things, that`d be 3 pins, one fo reach channel. Then i`d need to drive 4 bits to the driver,I could use the same 3 pins plus another for the last bit and one to control the multiplexer (CD4053). Tally so far : 5 pins
I`m using PIC16F675 (Quite a neat chip, 8 pins and it can do a lot of shit). This is my first PIC project.

Now I need to multiplex the displays, but with the remaining 6th pin I can only change between two of them

>:[

[Anonymous No.1505523]

>>1505521
Three things* (and three displays)

[Anonymous No.1505525]

>>1505500
Numerical integration. AKA a sum

[Anonymous No.1505563]

>>1505502
I just meant the appropriate box ie wall plate in the living room. Pic kinda related.
>>1505506
Will do, thanks!

[Anonymous No.1505564]

>>1505521

[Anonymous No.1505576]

Fuck. Finally. I have the power averaging working.
Fuck that pigeon fucker tesla and his shitty AC. Fucking curves, they make everything super complicated.
Edison had the right idea. Nice flat line with stable voltage, no autistic flinging it around up and down constantly, so easy and comfy to use and measure.

[Anonymous No.1505592]

R8 my car amplifier (preamp components are on the back side of the PCB). I haven't soldered on OEM amplifier connector and mounted the heatsinks yet.
>uses balanced inputs from factory radio
>3-way 4th order active crossover
>3 TPA3116D2 amplifiers for 2 tweeters and 4 woofers (bi amped is good)
>14.4V to 20V boost converter
>all on a 100mm x 100mm 2 layer PCB

[Anonymous No.1505602]

>>1505506
>>1505502
Here's the other side of the phone company's box. Really, unless the other 2 wires are also internet, I can't imagine the colors matter. Her used a red and green jumper to tie their blue/blue-white into the 'bus bar' which my house is connected to, then left the orange/orange-white.

[Anonymous No.1505604]

>>1505576
What problems you encountered? AC power in that scale has nothing complicated. Post piks

[Anonymous No.1505611]

>>1505602
From what I'm gathering, the incoming blue is data and orange is phone, so the box should probably be ruled out for lack of speed, along with you guys ID-ing the CAT 5E cable earlier.
I'll focus on hardware and the connection to the router next.
/blog.

[Anonymous No.1505612]

Elektro retard here, why are my resistors overheating? I think I did everything right.
AC 220 V
Resistors are 4.8k each
LED is idk what, 2.5 V?

[Anonymous No.1505613]

i have one of those small infra red light detecting 3 legged thingies, can it also detect normal light? from a led bulb for example?

[Anonymous No.1505618]

>>1505612
Well some basic math here (220V - 2.5V)/9.6k gives a current of about 22mA. Each resistor has 108.75V across it and a current of 22mA through them. Multiply those and the power dissipated in each resistor is about 2.4W so no shit they're burning up.

[Anonymous No.1505619]

>>1505612
nigger don`t mess with AC if you don`t know the basics of DC and AC power.
W=I*V

[Anonymous No.1505620]

>>1505619
also a meter ideally consumes/interfere as little as possible on what it`s measuring, so using resistors to measure mains with 1W precision monthly is kinda pointless.

[Anonymous No.1505621]

>>1505613
it's probably an IR signal receiver that only receives 38kHz modulated IR beams. if you want a visible light detector, use a phototransistor

[Anonymous No.1505624]

>>1505619
>>1505618
I get it. Resistors are built to handle only so much power and these watts are too much. How would I go about solving this? I just need to power the LED from 220.

[Anonymous No.1505627]

>>1505624
Bigger resistance or large power resistors or use a method that draws less current.

[Anonymous No.1505630]

>>1505624
First off use a high efficiency LED that can achieve good brightness at a relatively low current and then use a capacitive dropper for the current limiting. Use a 10nF cap for 1mA LED current or 100nF for about 10mA.

[Anonymous No.1505631]

>>1505627
You can buy ferrite cores or make someshit up with iron/steel wire or what ever and try, atleast it won`t be a fire hazard.

[Anonymous No.1505632]

>>1505631
Old charges and not utter shit chinkshit stuff have little rubberized cilinders on the wires, you can cut them up and use the ferrite.. You can use this for air cores.
https://en.wikipedia.org/wiki/Rogowski_coil

[Anonymous No.1505634]

>>1505612
>not even protecting LED from other half of sine wave
Might as well just electrocute yourself with AC right now

[Anonymous No.1505635]

>>1505621
I have a photoREsistor can i use that?

[Anonymous No.1505637]

>>1505635
yes. If you want it in some light flux unit you will have to calibrate, but for just sensing light it`s ok.

[Anonymous No.1505638]

>>1505635
did you try google?

[Anonymous No.1505647]

>>1505592
Not bad, but if you're putting a heat-sink on those TO-220s I hope you'll give it some affixation to the case.

[Anonymous No.1505653]

Hey guys I want to build a small to device to prank my co worker(s) at work

>I build cranes so its nosiy place

I want to build a device where I can hook it up to the horn, and either use radio frequency or possibly my smartphone (with esp8266) and set the horn off, and freak my co worker(s) out.

so the horn is a 12volt, and is a simple push button (I am planning to connect with the horn), with two eyelet connected to it. I was thinking of using a relay.

But most of my projects are either 3.3 volts to 5 volts. Is there such thing as a adjustable voltage relay? Where I can open/close the relay with 3-5 volts, and let 12v current through the relay?

Or are all relays ran on same voltage?

[Anonymous No.1505656]

>>1505653
contact voltage and coil voltage are independent and sually specified separately

[Anonymous No.1505658]

>>1505653
A normal 5V relay shouldn't have any problem holding back 12V on the switch contacts, there's at least a couple of mm separation there. 8266 through wifi sounds like a good idea, but an ESP32 with bluetooth is also an option. Not to mention standalone wifi/bluetooth shields for a minimal 8-pin micro. Also I'd advise using a common MOSFET instead of a relay since you'll probably need a transistor for the relay's coil anyhow.

[Anonymous No.1505695]

>>1505647
I originally intended to use a different cooler but it ended up being far too tall. I'm probably going to end up going with pic related (a northbridge cooler) and 3d print a mounting bracket and possibly a fan shrould so I can use a larger, quieter fan mounted like so.

[Anonymous No.1505717]

>>1505695
Sounds fine.

[Anonymous No.1505776]

>>1505592
>>1505695
pretty damn good if you designed that pcb , did you solder everything on yourself? Or can you get a the factory to do some of it for you. Tell us more.

t: im looking into getting some of my own amp pcb's printed off.

[Anonymous No.1505780]

>>1505602
The cat5e is nearly brand new, as is your demarc. Just call your provider. They can run a line test from the dsl equipment

[Anonymous No.1505786]

>>1503466
Help Share this Video Pleasehttps://youtu.be/xso2oo35bYA

[Anonymous No.1505806]

>>1505786
>switch from 4chan to 4channel
>suddenly far more referral link shills
what hath god wrought

[Anonymous No.1505824]

>>1505441
It's not crazy to think it could be. You'll get a lot of crosstalk from that mess and as the data integrity gets worse your devices communicate slower.

[Anonymous No.1505826]

>>1505634
It's protected by the 10k resistance bro

[Anonymous No.1505849]

>>1505826
Sure, if you want to put 20mA through it in reverse.

[Anonymous No.1505851]

>>1505851

[Anonymous No.1505870]

I watched a video where the man shown using a led with mains with a diode and 220k resistor in series, but 348/0.02 is only about 17k ohms i dont understand

[Anonymous No.1505872]

>>1505870
Nothing says you have to run the LED at 20mA or anything near to it. It depends on the LED how much light 20mA produces or is that much even allowed.

[Anonymous No.1505873]

>>1505872
This, running it at 20mA means dumping 2-5W through your dropper resistor. It's more efficient to use a neon since they run at 90V.

[Anonymous No.1505874]

To run an LED off mains with a 10k resistor just use a 5W resistor. Then it won't die!

[Anonymous No.1505884]

I don't really understand electrics but i have a question.
Power company guy showed up at my house to remove my old power meter and put in a new one.
After he installed it i noticed that instead of kwh the display shows "kwa" and when i asked about it, he said that the power company is switching billing from kwh to kwa
so what the fuck is kwa? is it same as kwh?

[Anonymous No.1505887]

>>1505884
Kwa or kva?

[Anonymous No.1505891]

>>1505884
"a" might stand for apparent, but that doesn't mean it doesn't need an "h" for hour.

[Anonymous No.1505892]

>>1505887
Not sure, i am at work right now, it's possible it's kva tho

[Anonymous No.1505898]

>>1505892
I've never seen that. Can you post a photo of the new meter when you're back? They cannot bill 'apparent' energy without legal trouble. Maybe they switch to yearly billing? If so, 1 kWa = 8760 kWh.

[Anonymous No.1505912]

So, normally a heat meter measures heat with two thermistor temperature probes and a flow sensor, so heat goes in, heat goes out and how much water it took and then it calculates the temperature diferences.
But i see they sell clamp on ultrasound heat meters now, pic related
how the fuck does that work? How can they measure heat and water flow without putting anything inside of the pipe?

[Anonymous No.1505915]

>>1505912
seeing they have two sensors in tandem, probably it has to do with wave propagation and wave speed and it`s relation to fluid flow.

[Anonymous No.1505916]

>>1505915
https://en.wikipedia.org/wiki/Ultrasonic_flow_meter

[Anonymous No.1505927]

>>1505916
>>1505915
that is pretty stupid.
if you like to listen to music, then the more you crank up the volume, the more will your heating bill raise

[Anonymous No.1505930]

>>1505927
Are you retarded?

Can anyon redpill me on re-using DVD lasers? I got four of them here, they all have three terminals. The internet tells me one of them is a photodiode for feedback control, but the internet also tells me the only people reusing them are indians with LM317 to control current. Anyone has experience with them? They have no part number so I can`t find specs to design a control system.

[Anonymous No.1505931]

>>1505930
I`ve got two from a R/W drive, so one of them is IR, but the lens that is glued to it cracked during removal. Godammed epoxy)

[Anonymous No.1505954]

>>1505849
If it can handle 20mA forward it can handle 20mA reverse

[Anonymous No.1505955]

>>1505954
Are you trying to be funny or something?

[Anonymous No.1505956]

>>1505927
turn your treble down and don't worry about it

>>1505954
that's not what your mother said

[Anonymous No.1505957]

>>1505955
Do you think semiconductors just magically break if you run them reverse biased? What do you think a Zener diode is?

[Anonymous No.1505959]

>>1505957
There's nothing magical in breaking shit with excessive power dissipation. Which is likely, considering the reverse breakdown voltage is much higher than the forward voltage drop.
Also, zeners are built so that the breakdown happens relatively evenly across the chip, instead of one small hot spot.

[Anonymous No.1505962]

>>1505959
It's barely twice as much power in reverse. If he designed his circuit to be at a reasonable power in the forward region he'd be fine in reverse.

[Anonymous No.1505967]

Can i run a piezzo buzzer directly from mains AC with just a resistor? i just want it to make a sound when i stick it into an outlet

[Anonymous No.1505968]

>>1505967
You need to look up the term "frequency response" and work out how it relates to a piezo.

[Anonymous No.1505970]

>>1505962
>It's barely twice as much power in reverse.
Where did that come from? Are you quoting some datasheet's absolute maximum value for the reverse voltage? It's not the guaranteed breakdown voltage and the actual breakdown voltage can be several tens of volts.
Relying on components to be better than their datasheets say to save one diode is dumb jewery and recommending that practice is even dumber.

[Anonymous No.1505974]

>>1505931
pls

[Anonymous No.1505975]

>>1505968
So no? Eh fuck it, i'll just use a led instead it will be easier

[Anonymous No.1505978]

>>1505975
You realise these are a thing, right?

[Anonymous No.1505981]

>>1505978
hahaha holy shit, only a complete autistic retard would EVER rely on these to test a live wire

[Bepis !n/ZnkB9jUw No.1505988]

>>1505981
Wew, they make those for 120V? That’s a lot of power going into a little bit of Chinesium

[Anonymous No.1506002]

>>1505408
I have been using ftp://www.zimmers.net/pub/cbm/schematics/computers/pet/index.html as a resource, that doesn't tell me anything that I don't already know, but thank you I guess.

>>1505410
> The term for that would be "cross reference"
Doesn't yield many more results, I guess substitution guides are a thing of the past.
> they're quite likely original.
The rust was a give away on this, hence my questioning about failure modes. Anyway, I dragged them out and tested them, they are dead short between their input and output pins, so I guess this board is fucked. Into the trash it goes.

>>1505967
Pic related, though a non contact tester will only test if power is present. A neon tester will let you test individual lines so you can tell if your wiring was installed by a mexican and the neutral is being switched instead of hot etc. You prolly don't want to be fucking with something that's home brew if you think wiring a piezo to a mains outlet is a tenable idea.

>>1505981
Yessiree, a far worse option than wiring a piezo to mains via a resistor.

>>1505988
You'd know all about chinesium wouldn't you, you vapid fuck.

[Anonymous No.1506054]

>>1506002
not so fast. plenty of that board might be NMOS. even the TTL might have survived if power supply fusing blew quickly enough. your chances are even better if the regulators were dead short between in/out and ground. it wouldn't be too much trouble to hook an ATX PSU + a 9Vac timebase up to it for a minute and see if anything still works, surely?

[Anonymous No.1506067]

>>1506002
>a non contact tester
is trash.
only a someone with extra chromosomes would trust their life to a non contact pen
the only correct way to test a live wire is with something that actually physically connects it to a ground wire such as a multimeter

[Bepis !n/ZnkB9jUw No.1506072]

>>1506002
>tfw no Fluke NCV

And yeah, I’m familiar with Chinesium. I am an American consumer so it would be difficult to stay away from it. I prefer Taiwanesium.

Anyway life and the internet have kinda taught me not to go as cheap as possible on testing equipment. I may not have $3000 worth of Fluke meters but ar least I didn’t go and buy some AliExpress “””True””” RMS meter.

I still gotta get me an NCV tester but I normally have the meter nearby when I fux wit electrons so I can stick the clamp in there and look for the little red light.

[Anonymous No.1506094]

every electrician has a fluke

[Anonymous No.1506097]

>>1506094
fluke, indeed

[Anonymous No.1506188]

Is Make:Electronics a good book?

[Anonymous No.1506208]

>>1506188
Download and see, buy if it suits you.
"Along the way, you will make some mistakes. This is good. Mistakes are the best of all learning processes. I want you to burn things out and mess things up, because this is how you learn the limits of components and materials."

[Anonymous No.1506211]

Can i power the tiny esp01 board from mains without using ugly and big power bricks?
What is some tiny little thing that can devour 230V AC and shit out 5V dc?

[Anonymous No.1506213]

>>1506211
20x34x15mm

[Anonymous No.1506214]

>>1506213

output 5v/3w

who the fuck specs a supply in terms of watts. I know the math is simple, but jeez.

[Anonymous No.1506216]

>>1506214
power supply designers who offer multiple voltage options in a single form factor, that's who
also chinks

[Anonymous No.1506219]

https://www.amazon.com/Fluke-Voltage-Detector-1000V-AC/dp/B004I9J4DI?th=1

[Anonymous No.1506220]

>>1506216

so are you with me or agin me?

who the fuck ever looked for a supply at a particular voltage and cared about watts, rather than amps.

>let the autism flow

[Anonymous No.1506221]

>>1506214
the first remarkable thing I noticed was the super generous 5mm separation of the high voltage ac inputs vs. the 15mm separation of the low voltage dc outputs
so I don't recommend you use this in your showercam

[Anonymous No.1506225]

>>1506220
probably very few people. otoh people who looked for a supply to fit a particular volume then worried about voltage probably thought out their watts first, because voltages are fairly easy to change
>pic
roll

[Anonymous No.1506253]

>>1503521
Ever heard of a stool

[Anonymous No.1506273]

>>1506214
i will blame audiofags, transformers are always given by power determining physical size, probably power matching speaker/amp power i guess...

[Anonymous No.1506284]

>>1506253
one of them magic stools where your legs dissapear. fucking idiot.

[Anonymous No.1506353]

>>1506211
Tear apart a phone charger

[Anonymous No.1506362]

>>1506211
search "AC 5V buck" on ali and you'll find some pretty tiny modules. If you don't mind wasting half your power again in a low-dropout linear reg you could just use a dropper capacitor > rectifier > regulator dealio, which is arguably more efficient if you're going really low-power.

[Anonymous No.1506406]

>>1506213
>20x34x15mm
that is fucking massive
look how tiny this thing is compared to even my midget hand
I can can literally fit it under outlet covers without any drilling and shit, so adding a 3cm brick to power it would really suck a lot
>>1506353
too massive
>>1506362
>AC 5V buck
i already own several of those with the small transformers ones, but they are still fuckhuge compared to the esp01, i don't want the powersource be 4x larger than the powered device.
Am i basically fucked here?
>dropper capacitor > rectifier > regulator dealio
i don't care about wasting power since it will be powered by mains, even if it is wasting like 100ma that's nothing, as long as it won't set my house on fire
do you have circuit?

[Anonymous No.1506410]

>>1506406
Here you are:
Basically you'll need to define a maximum current and be prepared to always run near that. You'll select a capacitor value such that this current causes its reactance [ X = 1/(2πf*C) ] to be enough to drop the voltage across the rectifier to a volt or so above the 5V/3.3V you need. It's reactive power, so the calculation isn't very straightforward, so I'd just test it in spice (free) until it works as you need it to. The minimum voltage across the rectifier needs to be at least the minimum regulator dropout and appropriate rectifier drop higher than what you want across the ESP. Ensure the dropper cap is not polarised (a film capacitor should do) and can handle the full peak voltage of your mains. Test with a resistor instead of your ESP01 for obvious reasons.

Things to possibly add:
-resistors in parallel with large capacitors to prevent them staying charged up
-thermal fuse on the regulator just in case it gets too hot
-full bridge rectifier for a smaller filter cap

Perhaps look at some of Big Clive's cheaper LED bulb teardowns as they often contain such a power supply for the LEDs, where the capacitor is simply acting as the current limiting resistor. Since the current through a capacitor is 90° out of phase with the voltage, the net power consumed by the capacitor is zero; it simply collects power and feeds it back into the wall again instead of heating up like a resistor does. An inductor can do the same thing, but getting a big enough inductor would take up a lot of room.

[Anonymous No.1506411]

>>1506410
So this then eh?
https://www.youtube.com/watch?v=SIB-cBPaDf4

[Anonymous No.1506412]

>>1506411
Looks about right, yes. Though testing one in spice it seems you probably have to use a full-bridge.

[Anonymous No.1506414]

>>1506412
Oh and a normal regulator doesn't seem to shunt the input voltage to ground if it gets too high, so either just use a zener or use a zener before the LDO to clip the voltage. I wouldn't recommend just running an MCU off a zener without an LDO, but it wouldn't be the end of the world.

Pic related looks to work fine on 240VAC up to 5mA output, and even at that it goes though bugger-all power. The 100µF cap could probably be smaller too.

[Anonymous No.1506415]

Someone recommend me an 8-to-3 encoder, I've been googling for two hours now and can only find priority encoders. I swear I'm going to develop brain hernia at this rate

[Anonymous No.1506428]

>>1506415
What kind of behavior you're expecting? Output = number of input pins high?

[Anonymous No.1506432]

>>1506428
>Output = number of input pins high?
Yes, just a octo to binary conversion. It seems like no matter what I try to search for AI only get priority encoders who only encode the highest pin

https://en.m.wikipedia.org/wiki/Encoder_(digital)

[Anonymous No.1506434]

I have a newbie question, I'm working on a game boy project with a game that was designed to use a custom token ring to enable more than two players to link together. As I understand it, a token ring is a network architecture where a piece of data is passed from the output terminal of game boy A to the input terminal of game boy A+1, all the way to game boy A+n, which feeds its output back to game boy A, completing a loop.

I have verified that this system does in fact work for this game, but I am having trouble getting connections to be working (let alone stable) past the point of 10 game boys or so. As I understand it, the game was intended to do 16 players, but I am having a hell of a time getting that to work. It seems like the token keeps getting passed around, and then when you add that many game boys, it just can't make the loop around the whole path.

I guess my concern is that maybe the hardware I'm using isn't up to snuff. I'm using some cheap chinese game boy advance cables that I bought in bulk off of ebay. They seem to work fine up to the 10 player limit that I explained, but then fail to connect much further. I remember hearing once about signal drops in cables, could that be the cause of the problems that I'm seeing? I could imagine that the game boy just doesn't have enough power to push data across that much cable (each link of the cable ring is about 12-14" and there are 10+ of them, Nintendo never made a game boy link longer than 3 feet in length). That said, I don't know a lot about this subject, so I return to /ohm/ to be enlightened by you fine folks. If anyone can weigh in and offer advice, then I'd be very thankful for it. Peace and love, friends.

[Anonymous No.1506450]

>>1506432
Have a look at the truth tables on the article there, an n input priority encoder does the exact same thing as an n+1 input generic encoder for the case of a single input high, and adds extra valid outputs for 2+ inputs high where the generic encoder would be undefined.

[Anonymous No.1506462]

>>1506450
I'll take a look when I get home, it's been a few years since I made any significant board design and I might just have overlooked something

[Anonymous No.1506464]

>>1506406
if you're running an ESP8266 you'll need 200mA or more output capability, i.e. 1W+ plus whatever you're trying to power with it
I see 25x25x12mm converters but they don't get much smaller than that
you might also have a look at
https://www.microchip.com/ParamChartSearch/chart.aspx?branchID=90251
and be prepared to roll your own. Pic potentially related

>>1506432
so you want an 8-wide 1-bit adder which is not an encoder
pretty sure there's no one-chip solution for that unless you build your own in a PAL or CPLD

[Anonymous No.1506480]

>>1506464
I'll look into it, it's possible that I'm just confusing encoders with something else

[Anonymous No.1506526]

>wire up chip
>connect red wire to 12v black wire to ground yellow to data
>why is my shit getting so hot
>it was because red is ground and black is vcc
fuck you chiiiiiiiiiiiiiiiiinks

[Anonymous No.1506527]

>>1506526
qaulity

[Anonymous No.1506533]

>>1506406
Personally I'd play around with linkswitch-tn2 or viper series AC/DC ICs in non-isolated buck topology to get required current power output with minimal effort. Capacitive power supplies are shit, to get more current from them they begin to increase in size significantly (because x caps are huge).

>>1506526
>company ordered optocoupler drivers for some obscure use
>one output is usb cable, another is db9
>db9 soldered in reverse
>for usb - white is 5V, red is ground, black is d+, all soldered incorrectly
At least they are recycling (by using discarded USB cables for stupid westerners).

[Anonymous No.1506545]

>>1506533
>soldering tiny laser
>solder blob goes down one of the legs, somehow spreads over the whole flat part shorting all legs
>can`t suck the solder thin layer
>wicker doesn`t work
>tried scraping it with an X-knife, no dice
>tfw going to have to throw 2 IR lasers away

[Anonymous No.1506548]

>>1506410
Test behind isolation transformer for obvious reasons.
>>1506414
Same. Fortunately no one will built it.

[Anonymous No.1506565]

>tfw thing stops working AFTER you put all the heat shrinks on
fug

[Anonymous No.1506616]

Is it just me or has /ohm/ been even more retarded than usual lately?

[Anonymous No.1506666]

>>1506616
not just you

>>1506533
thanks for nudging me to investigate the VIPers a bit more. the few versions that can do non-isolated buck don't look like they can do less than about 10.5Vout so OP would have to follow up with a post-reg stage, preferably in his case one that conserves voltage as current instead of just pissing it away as heat

[Anonymous No.1506705]

By accident I`ve ordered 4 old ass twenty something PICs (PIC16C620), the oldbois only have a buttload of pins, and no ADC. My idea is to use one or two of them to measure 0-20v off my psu and output that to 7 segment displays, making 2 ghetto voltmeters. The way I`m thinking of doing that is comparing voltages while stepping down the reference (start at the largest and go to the smaller ones). Has anyone tried something similar? Using that the largest resolution would be 0.8V/increment which is kinda shitty but better than nothing.

[Anonymous No.1506706]

>>1506705
>twenty something
Twenty something pins*, and each of them have 2 comparators each.

[Anonymous No.1506711]

>>1506666
VIPer06 definitely is capable of 5V and less output at around 200mA.
Other in this series as well, VIPer26, VIPer115 etc. They might support higher continous output power.
Output voltage depends on feedback resistor values

Though I would probably go for low frequency ones (30kHz) and limit output current to about 100mA (enough to run ESP8266 without many issues provided sufficient bulk capacitance for wifi communications) as at 4V or 5V output efficiency of ac/dc buck converters isn't exactly spectacular. But for 1-2W designs I'd say just go for it and test (though use of isolation transformer on first iterations is always a good practice to keep).
Still more useful than 20mA current of capacitive supplies.

ST provides a rather good tool for initial design for buck converters to get you started on design.
But for >100mA outputs I'd watch thermal performance of SMPS ICs, maybe use bigger copper plane for pcb layout or some heatsinking.

[Anonymous No.1506716]

>>1506705
as far as I'm aware feeding 20V into PIC will kill it. Unless I'm mistaking something but Vref should not be much higher than chip supply voltage.

Why not use op-amp in comparator mode? One input could be resistor ladder to get desired voltage (DAC) and another would be voltage input from PSU? (or just get serial ADC)

[Anonymous No.1506717]

>>1506705
a ramp ADC would be faster and possibly simpler, and give you as many digits as you care to wait for

>>1506711
I was misled by a lack of a sample design in the VIPer26 datasheet and/or bad fonts. so, cool
dat turn-on spike doe. I'd put the current limit up a bit higher than that

[Anonymous No.1506722]

>>1506716
>>1506717
Yeah I`m retarded, I`ll feed it off a 5v rail. If I use a voltage divider to feed the input voltage I`ll have a 0.2v resolution. (Scaling down 20V to 5*15/24 = 3.125V)
>>1506717
Care to explain? They don`t have ADCs or DACs in them.
>>1506716
I`m trying to use the pics, I bought them by accident and they are already shipped.

[Anonymous No.1506729]

>>1506722
>They don`t have ADCs
Basically you use the PIC to measure the time it takes for a capacitor to charge in such a way that this time is proportional to the voltage you're trying to measure. I think it requires some sample+hold wackiness but nothing too obtuse. Of course the actual voltage precision is dependant on the part values and possibly noise suppression.

[Anonymous No.1506731]

>>1506729
Ah, nice.
>I think it requires some sample+hold wackiness
I don`t think so, just some interrupts and a comparator interrupt.

[Anonymous No.1506744]

>>1506731
If that's the case then no problem.
>using PICs in 2018

[Anonymous No.1506746]

>>1506744
what is the problem with PICs?

[Anonymous No.1506762]

>>1506731
you will probably want to use an op-amp integrator as your ramp generator, since the PIC wants a low-Z source for its analog inputs. Pic related, clock is ~95% duty cycle, 1V offset is to allow the (negating) integrator to zero. there's probably a better way to do that but I can't be arsed on a friday afternoon

>>1506746
4 cycles/instruction, for one thing. the W register, for another
I grant they are slightly cheaper than AVR and have slightly better peripheral options, sometimes

[Anonymous No.1506781]

Redpill me on ssrs. They make mech relays seem like untermensch in every way

[Anonymous No.1506798]

>>1506762
Yeah I was thinking about the op-amp. I have some spares atmega328 but they are expensive compared to a 1$ pic, I`m not doing anything super fast, I think the fastest think I`ll do will be related to SMPS or a pic recording data and sending it to a radio transceiver. (A remote sensor)

[Anonymous No.1506799]

>>1506781
Insulation, resistance, it usually fails short and if you need to switch something very fast using relays doesn`t seem smart. There are other devices made for that. e.g IGBT, thrystors, triacs.. Ah, and price.

[Anonymous No.1506823]

>>1506798
>expensive compared to a 1$ pic
ATtiny84 (14 pin) likely has higher specs than most PICs and cost perhaps $1.6 each (SMT in a 10 pack from Ali). ATtiny85s are even cheaper, though only have 8 pins.

[Anonymous No.1506829]

>>1506823
If you are american and can buy things from china. Mail services here in brazil are crazy expensive and now the jews are charging 15R$ (a regular lunch) for any foreign package, regardless of tax and size. Anything remotely connected with this new "maker"" bullshit, arduinos and ATtinys included, are 4x the price just because BRs are jews.

So I started using PICs for stuff because they are somewhat common, only some stupid techs that treat them like voodo buy them so stores usually have a stock of old and new ones.

I`m hoping I`ll find a lucky store in ali with very cheap nanos+other components I need so I can buy a lot of shit, just below the tax limit and only pay for one package.

End of rant

[Anonymous No.1506832]

>>1506829
>american
I'm in Australia. Ali gives you free shipping to basically fucking everywhere, I'm sure I've seen Brazilians on >>>/g/csg. But instead of going for the absolutely cheapest sellers, go for ones with a substantial amount of purchasers. The Aliprice browser plugin helps greatly.

[Anonymous No.1506835]

>>1506832
No free shipping here buddy. 15 br monies for ANY package that comes in. And mail service stated on several ocasions that chinkshit have 0% priority so it takes 4 months to get anything..
>aussie
Anglo, whatever, also you are almost chinese by now so shipping is cheap.

[Anonymous No.1506839]

>>1506823
ali isn't necessarily the cheapest place to get AVRs without an arduino attached. sometimes digikey is cheaper. for example, right now they have ATtiny84s in 20VQFN for $0.40 each in singles, if you can deal with that small of a part. SO-14 is $.80 each in singles
compare to PIC16F15323 in SO-14 at $0.71 each in singles

>>1506832
>go for ones with a substantial amount of purchasers
this anon fucks

>>1506835
looks like there are a few full-line distributors in Sao Paulo. they would be quicker, if not cheaper
>almost chinese
kek

[Anonymous No.1506843]

>>1506839
yeah, Sao paulo is great, it is world class in the area of finding shit you need, it has literally anything, a shame I only go there once a year, and again shipping here is retarded because of statal monopoly, shipping things from china to here is cheaper than sending a package two towns (lets say the average IC costs 0.6 br money, shipping to here, and I live in a capital, usually starts at 25 monkey moneys). But thanks for the tip guys. And the only good part about all this shit is that working with these constraints really get your noodles warm.

[Anonymous No.1506845]

>>1506835
>15 br monies for ANY package that comes in
That is very sad, but at least you aren't Venezuela.
>you are almost chinese by now so shipping is cheap
They're not intrinsically related, we've just got trade agreements.

>>1506839
>sometimes digikey is cheaper
If this is the case and they have a warehouse in your country (Digikey, Mouser, Arrow, etc.), it might be worth getting a bunch of shit from there and riding on the "sufficient purchase = free shipping" dealio. I think Arrow have the lowest price limit for getting free shipping, at least in the western world.

[Anonymous No.1506927]

Anybody have any experience with FPAAs? Are they actually useful or are they a bit of a meme?

[Anonymous No.1506931]

>>1503466
Question I can't find the answer to because it is too specific. If I buy a 100Ah 12v lead acid battery, should I go ahead and treat it like a 50Ah battery because most lead acids can't be discharged below 50%? Do the manufacturers take this into consideration when labeling their batteries by halving the usable amp hours on the label?

[Anonymous No.1506943]

>>1506931
If it's a deep-cycle battery, probably not.If it's a car battery, who knows? Look for the datasheet in any case, and see if there's an I/V curve on it from which you could estimate the capacity for intended use.

[Anonymous No.1506969]

>>1506927
You mean FPAA instead of FPGA? Just looked up an FPAA, actually pretty interesting looking but now-days with FFTs and fast digital logic in an MCU I can't imagine anyone using one, except possibly for radio frequency purposes.

[Anonymous No.1507042]

>>1506799
Chinks sell them for under one buck and they isolate completely that is their whole purpose as a relay

[Anonymous No.1507061]

>>1507042
>they isolate completely
They have an insulated gate, sure, but it's only a some dozens of microns of SiO2 compared to ~5mm in a real relay. And QC and warranties from $1 noname chinkshit that needs a whopping great heat sink is a deal-breaker for me any day.

[Anonymous No.1507062]

>>1507042
>reverse current
>isolated

[Anonymous No.1507064]

>>1507062
Also if you are buying 1$ relays, who the fuck cares, it`s just a low voltage, low current thing. The guy that asked the question didn`t say at what power levels it`d be used or in what conditions.

[Anonymous No.1507073]

>>1507064
>low voltage, low current thing
Then use a damn MOSFET. Only reason to use an SSR over an appropriately rates FET or IGBT is if it's running on AC. If you're switching low-voltage low-current AC then sure use an SSR but that's a very specific use case.

[Anonymous No.1507076]

>>1507073
Hey dude, I`m not the guy asking or defending SSRs, every equipement has its use.

[Anonymous No.1507078]

Fresh bread.
>>1507025
>>1507025
>>1507025