/diy/ - Do It Yourself

>>1591072
you forgot the OP. fucking christ

thread with indefinite lead time: >>1586417 (Cross-thread)

>RULES
0. Electrics ≠ electronics. Home appliances/sparkies to /qtddtot/ or /sqt/. PC assembly to >>>/g/.
1. Do your own homework. Search web first. Re-read all documentation/datasheets related to your components/circuits. THEN ask.
2. Pics > 1000 words. Post relevant 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.
4. /ohm/ is an anonymous, non-smoking general.

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

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

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

>Design/verification tools:
LTSpice
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)
Local independent electronics distributors
ladyada.net/library/procure/hobbyist.html

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

OP is stupid

Thread from the bone zone:
>>1591036

>>1591073
There's always the randomiser+buzzer or speaker that you hide behind a refrigerator, which is usually done with an MCU but a LFSR would work also.
Things that shock people could also be funny xd, but there's a chance that they could damage sensitive electronics so I'd be careful with that.
Fake cellphone ringtones would be pretty easy to do.
USB killer?
Hooking up a battery-powered bluetooth receiver or MP3 player inside someone's stereo so even when they unplug it they can still hear Brenton Tarrant's playlist could be pretty xdxdlmao, eh?

you fucked up even worse than the other op

I accidentally destroyed a specific capacitor on a board (pic relate) and so following some advice I looked up the datasheet to try to find out what capacitor the related chip would recommend...

How do I find that anywhere in this lengthy datasheet thats way over my head? There's 45 pages and I'm not seeing anything about capacitors on it anywhere for this little IDE to Sata controller chip (JM20330).

As a side question, how many of you can understand all of the complex jargon thats listed in one of these data sheets and how long did it take you to get to a point where you can read one fully and understand everything that's going on in one? I'd like to get to that level someday.

I accidentally destroyed a specific capacitor on a board (pic relate) and so following some advice I looked up the datasheet to try to find out what capacitor the related chip would recommend...

How do I find that anywhere in this lengthy datasheet thats way over my head? There's 45 pages and I'm not seeing anything about capacitors on it anywhere for this little IDE to Sata controller chip (JM20330).

As a side question, how many of you can understand all of the complex jargon thats listed in one of these data sheets and how long did it take you to get to a point where you can read one fully and understand everything that's going on in one? I'd like to get to that level someday.

>>1591158
ctrl+f "capacitor"
how many results
>inb4 scanned datasheet
shit out of luck in that case mate

>>1591161
"Capacitor" 0 results
"capacitance" 5 Results

Is that what I should be investigating?

>>1591166
Uhh, how about "nF" or "pF"? Looking for example circuits or application notes or whatever should be your priority however. In any case, check out those 5 results.

>>1591170
4 of those results are identical: "It should be bypassed to ground by a 0.1uF capacitance."

So I need to look for a replacement 0.1uF capacitor? I went to Aliexpress and found this and assume it's what I need to order:
https://www.aliexpress.com/item/100PCS-0603-104K-100nf-0-1uf-10-X7R-50V-chip-SMD-Ceramic-capacitor/32966946280.html

Am I correct? Also thank you.

>>1591173
>It should be bypassed to ground by a 0.1uF capacitance
I'd first check that one of the pins it's talking about is the one connected to C5, but considering there are 4 bypass capacitors on board, I'd say those are a pretty good guess.

>aliexpress.etc
>0603
I'd check that those are the right size, measure the size of the other bypass caps nearby and compare them to whatever "0603" actually measures. I think it's some negative power of ten of an inch or something.

>>1591173
>>1591178
Oh but before you buy anything, check the pads on the board, ripping the cap off may have delaminated the copper there, in which case buying such a replacement won't be terribly useful. I'd lightly reflow the two pads, perhaps with a piece of braid to clear off excess solder and debris with it. I trust you have some way of soldering a small part in place?

>>1591178
Thank you anon, will do.

>>1591179
I don't have a heat gun or anything like that for smd soldering but I have an iron and some braid.

It actually didn't seem to rip the pads off, but the plier tip I was using knocked right through the part and left the ends soldered on it seems.

>>1591182
Yes I guess it's a good thing that SMT components are easier to break than the pads they're soldered onto. So long as you have well-aligned tweezers that can get in there you should be fine. My tweezers are about 0.5-1mm off at the end (in both directions), so I won't be using them for SMT any time soon.

Anyone have a recommendation for a decently priced oscilloscope? I was looking at the Rigol DS1054Z after seeing Great Scott use one, but the "trial period" bullshit has me spooked

>>1591189
>trial period
Never heard about that, though I have heard about the hackable Rigols that people increase the bandwidth or something of.

>>1591189
>trial period
No such thing on my Rigol, but I haven't played around with any of the newer Z units.
>>1591192
Pic related.

>>1591200
There some resource on which scopes can be hacked? A second-hand one might be a pretty good buy.

>>1591204
I haven't looked into it beyond this one by accident. You literally feed it an update for the "higher" tier model and it just works because they have exactly the same hardware and only the factory loaded firmware and stickers differentiate them.

>>1591207
Oh that sounds pretty simple, it would just be a matter of figuring out which scopes have the same hardware. Are they easy to update?

>>1591209
In my case I just stuck a flash drive with the firmware on it into the usb port and it immediately prompted to install the update.
Apparently the Z series Rigols do have features software-locked behind serial keys (so I guess they could have the time-limited trials the other anon mentioned?) but they're hackable too so that ultimately doesn't matter:
https://www.youtube.com/watch?v=AnxOGl27GjE

>>>>1591042
Did you mean this? That's now obsolete because of the mental climate change.

>>1591213
You are stealing money from the manufacturer. This is the reason all comapnies hate hobbyists. They all all greedy jews who want to run them dry without paying a penny for their work.

>>1591158
I recognize that bulge
SATA lanes use 0.1uF capacitors as their dc blocking caps. size is 0603 iirc but you should a millimeter ruler or some digital calipers just to check

>>1591308
yet Microsoft was more than happy to have windows on lots of machines, even if they didn't get paid for every copy. now why might that be? network effects? free advertising? making 99% profit sales to corporates who aren't going to want to hack?
take your simple-minded moralfaggotry elsewhere

>>1591286
rollin'

Could it be that my hall sensor doesn't work when I try to hook it up to a 3.3V sensitive gpio pin like in pic related?
I think it should work like this but it doesn't.
Works perfectly fine without the voltage divider and a 5V gpio pin.
I think I should just put the output on a transistor base and switch a 3.3V rail on an off instead...

>>1591320
That's a bit odd looking, wouldn't the voltage only go up to 2.5V? That's borderline logical high there, I'd use a 1k and a 2k instead. If the thing has a push-pull output, it might have an extra 0.6V or so from 5V, pushing you down to 4.4V/2 = 2.2V which is even less likely to be a well-received logical high. Got a scope to probe that pin? How about posting the datasheet and its IO pin equivalent circuits?

>>1591320
why should it work?

>>1591320
>>1591326
>it's open collector
Use a fucking pull-up to 3.3V you shitter

>>1591326
Hm I read 2.5 is enough for he raspberry I'm using, due to paranoia I decided to go with 2 1k resistors
>Got a scope to probe that pin?
I could only measure the voltage when no magnet is nearby because I need two hands to hold my probes.
I forgot the datasheet
https://cdn-reichelt.de/documents/datenblatt/B400/TLE49X5L.pdf

>>1591328
Oh so I can't just pull up the output to 5V and then voltage divide it?

>>1591337
>pull up the output to 5V and then voltage divide it
You could, but the pullup and the divider itself would make another divider, there's no reason not to use a pullup resistor directly to 3.3V, since there should be a 3.3V rail around somewhere.

>>1591339
That makes sense, thanks

What's a recommended board for learning about microcontroller?
End goal is to control pwm output for a DC motor.
Ideally something friendly like an Arduino but 32 bits. Also Linux friendly.
I want to learn about interrupts and embedded systems.

Are those sbc what I'm looking for?

I find the st micro nucleo but it's Windows only.

>>1591372
>intel
discarded
just get a bluepill. you can program it with the lolduino IDE on linux just fine, or program closer to the metal using CMSIS or HAL libraries

>>1591372
wait, do you mean something you can run linux on?

>>1591372
>End goal is to control pwm output for a DC motor.
Why do you need 32 bit just for that?

what are the 'flattest' batteries that can deliver a good amount of power for a raspi? Im limited in space and 18650s are too cylindrical but something much flatter that can offer similar power would really help

>>1591389
The flattest batteries are LiPos.

>>1591389
I'd go with a replacement battery for phones or tablets.
They tend to be very flat and high capacity.

Hi guys, i need to make some simple standalone push led light (1 push = ON 1 push again = OFF)but have measure constraints.
They need to be cicular, around 5cm diam, and between 5-10 mm thick.
The main problem is the height of the mechanical push buttons.
So is was looking at diagram to do the on/off electronically. An eventually found the mini pushbutton boards by Pololu.
https://www.pololu.com/product/2808

Is it the smallest i can find?
I lack the proper english words to search an alternative.

Thanks

>>1591388
this

>>1591372
adding, STM32F103/bluepills have a very good timer that's perfect for all kinds of motor control

>>1591527
if you care about size constraints, modules aren't a very good choice, especially not ones like that with all those bells and whistles
for a first go I'd probably build my own board, with a TTP223 touch switch IC connected to a ring at the board perimeter as the touch electrode, with a MOSFET turning the LEDs on and off
tell us more about your present design constraints, production methods, skill level, etc.

>>1591532
>tell us more about your present design constraints, production methods, skill level, etc.

It's part of a furniture design for a yacht. I visualize small and thin discs. I was thinking about making a two part shell with 5-6 leds inside and a button cell battery.
A touch switch is a damn good idea! thanks for that.
I'm a woodworker but made several arduino stuff and used to work in the aeronautic field and have a diploma in industrial automatism.

I only need 5 of them, everything will be handcrafted

>>1591537
>button cell battery
if you want to replace them every three hours of usage, this is a great idea
are you sure you can't wire into cabin power?

>>1591189
That’s the one I have on my wishlist. I’m about to break up with my girl so I can impulse buy it.

It has great reviews, tons of people own them with very few complaints, and everybody does that hack which makes it the same as some model that costs like $100 more

Somewhat related, are there any affordable portable scopes that could replace a benchtop unit for a hobbyist and don’t suck? I know there’s the PC units too.

>>1591189
lmao, my local reseller is giving me the price as $540.

>>1591308
In this case it’s not terrible. It’s like free education, plus you’re still paying $400 for a machine. It makes equipment like that attainable for the average geek and you never know what good that geek can create when he/she would’ve stayed away from the hobby in the first place if it cost thousands to get started. That’s why Arduino and all of the other open source programs do it.

>>1591538
you have to be able to remove the discs at will. 3hours seems low, a bicycle small red light thing lives at least 24 hour. At first i was tempted to use one of those little light, some are really slim, but they always have blinking modes.

If i connect a battery in parallel with a wall adapter and the battery is say 4V and the adapater 5V, then if i put an ordinary diode on the + wire coming from the battery the battery will only start supplying power if the wall adapter turns off right? since front of the diode V needs to be lower than back of it by about .5V

>>1591550
they must be driving them rather lightly then
you should definitely test whatever LEDs you're going to use first with some various batteries, so you can make a choice that meets your brightness and lifetime requirements. the touch switch doesn't draw all that much current (3.0µA max) so the user might have to replace the batteries every several months even if not used. to get much less current consumption than that will require some microcontrollers and programming or some unusual parts I don't happen to know about

>>1591555
correct, but you probably want to put a diode on the adapter itself in case its regulation circuitry draws any back current

>>1591382
My computer runs on Linux and I don't want to install Windows for the store purpose of installing a toolkit
>>1591388
This is my first project. This goal is simple but I want to familiarize with a 32 bit microcontroller so I can try getting promoted to another team at work.

>>1591555
Requires two diodes, one from the adapter (+) and one from the battery (+), both pointing to the load, common (-).

>>1591593
ah okay, I see no reason you shouldn't be able to use the Nucleo just fine with the TrueSTUDIO IDE which is Java-based and runs on Linux pretty well
or you can just use the Java-based STM32CubeMX utility to generate skeleton projects, fill them in, and compile them with your choice of ARM gcc/binutils. this is what I do for my bluepill

Ok now I’m confused. That fuse was tied between the black and grey, but they were all soldered together anyway. How the fuck does that work? There must have been something between the black and grey that were tied together by the fuse.

Should I solder it all together in a bunch or separate them after I put the fuse between them? The second one seems like the logical approach.

>>1591606
if you can't put things back how you found them, maybe you should give up

>>1591608
I’m gonna try. It will either work or it won’t. Or it will work for a couple days and then burn down the house.

I blame Amazon’s shipping times. If the drone delivered the thing 10min after I took the fan apart and ordered the fuse, this would be easier.

>>1591606
>but they were all soldered together anyway
What do you mean, soldered together and shorting the fuse? If it's just one end that has a bunch of wires soldered to it, that's natural because it's neutral.

Also I wonder if a PTC thermistor (as usually used as resettable fuse) would work as a resettable thermal fuse?

>>1591606
lots of wire nuts

So, /ohm/, I just fell for the audiofool meme.

I have a pair of Audio Spectrum speakers (local builder using Seas units) that I killed when I was a kid. My dad send the drivers to be repaired but the repair was not very successful.

15 years later, I decided it's time to repair them to their former glory, so I found the same exact new-old-stock drivers and installed them. The new drivers provided an audible improvement, especially when it came to high frequency distortion. But visual examination of the crossover worried me so I decided to replace all caps and resistors on it to be sure.

>>1591651
Turns out i was memed, as the original caps and resistors still measured well and subjectively, the speakers sound the same still.

Oh well.

>>1591652
To be honest, I'd be more worried about the melty inductors, surely they're not supposed to do that?

>>1591625
Yeah it looked like the black, grey, and both ends of the fuse were all soldered in one bunch. But I have a feeling maybe there was a tiny little bit dividing them.

Anyway she runs now. I may have fucked up the bushing deal for the shaft but I guess we’ll see how long she goes. I have 9 more fuses.

>>1591608
And this is a response from somebody who has no answer but will act like they know all.

>>1591653
The plastic form melted due to high temperatures (I was dumb when young and drove them at clipping for a long time)but the inductors themselves are fine. The magnet wire has not discolored anywhere and the shape has not changed, so they are effectively undamaged.

>>1591657
>drove them at clipping for a long time
lmao, doing it for the volume or the kind of sound it makes? I should make a guitar pedal that's made of an amplifier and attenuator, with a double gang pot alternately controlling the gain/attenuation of both, so the output amplitude is the same but you can end up with clipping.

>>1591661
y u no just follow with a compressor? you'd probably want one anyway

>>1591671
I don't even know what that is, I've never worked with guitar pedals before nor do I have any idea of what constitutes a common guitar pedal.

>>1591672
dynamic range compressor. it amplifies signals but reduces amplification when the signal exceeds a settable threshold. can help reduce variations in volume between notes

>>1591072
Who is this hot dog fucker?!

>>1591689
Wait, do you mean "hot dog" "fucker" or "hot" "dog fucker"?

Can I use hanging picture wire to make solder bridges or will it mess up my circuit with too much resistance.

>>1591791
Just save the cut off legs from through-hole components and use those.

>>1591792
This. Solid cores from network cable or phone cable also works well, but they have to be solid core, not stranded core.

>>1591661
>lmao, doing it for the volume or the kind of sound it makes

I used to be a dumb 14 year-old.

>>1591655
If it runs you've done it right.
>9 more fuses
Kudos for your perseverance!

>>1591824
It ran all night on medium speed and hasn’t burnt up yet.

So on the fan like that, the speed settings, is that just changing the frequency of the AC wave?

Also I was kinda surprised that each speed setting it hooked up to one tiny strand of copper that goes to the winding. Seems like a lot of power through that little thread. But I guess it’s also 120V, so it’s probably <1A and if you were trying to run that same size motor off 12V DC it would burn up that little wire in a second.

>>1591893
more wires = less resistance = more current = more magnetic flux

>>1591906
No I’m talking about where the wires coming up from the switches terminate at the fan motor, the “L-M-H and N”, they have a single tiny little strand of copper soldered to them and that copper goes to the fan motor. So there’s like 1” where all of that power to the motor is traveling through one strand of copper.

Also I was soldering and then using a little butane torch for heat shrink and I realized I was about half an inch from a thermal fuse that blows at 130F. I’m glad I remembered that small detail before I fucked everything up. Checked for continuity between the fuse leads like 10 different times.

What can cause a multimeter to read more resistance in a circuit than "actually present", going by the resistors I'm measuring?
I have a dpot driving a voltage step-up module, and when I measure the resistance on the dpot, it gives me like 50k, when it should be around 2-3k (and it is with the step-up module disconnected). There's also a 4.7k regular .5W resistor over the two pins of the dpot and it's not getting hot, I would've thought such a setup could never register more than 4.7k, only less.

>>1591943
a capacitor

I don't understand how connecting two non isolated power sources in parallel can cause a short but in series it doesn't?
For example two computer power supplies
And why cutting the green wire magically solves that problem.
It's just as if people are just making those rules up.

>>1591970
look inside the power supplies themselves to see why that is often the case

>>1591893
>So on the fan like that, the speed settings, is that just changing the frequency of the AC wave?
If you don't like Google, there's Bing, DuckDuckGo, and other search engines.
(pic relate filename is search term)

>>1591970
>connecting two non isolated power sources in parallel can cause a short but in series it doesn't?

I wouldn't connect two supplies in parallel, but that would not cause a short assuming they are the same voltage. Any two in series would short the one that has ground on both ends.

>>1591893
An induction motor resembles a transformer with primary windings as stator and a moving, shorted secondary 'winding', the iron rotor. Induction motors are asynchronous motors that run on permanent slip depending on power and torque. The load torque comes from the fan/propeller, which depends on speed, on how much air per time it displaces. The balance of of both determines the speed the fan runs at. Low, medium, high power. The no-load speed is the highest the motor can run at with very little slip. Or so I think. Wikipedia certainly has a much more accurate description.

Your fan motor is specified to max 55W, 120V, 60Hz. I know nothing about the power factor. If you have one of these energy meters that you can plug between mains socket and load you can read voltage, current, power and power factor cos(phi), the phase shift between mains voltage and current.

>>1591974
That pic doesn’t help. All 3 wires attach to the main winding and they all send 120V to the motor when it’s switched to that setting. So is the frequency changing? Or is it PWM type sheet?

>>1592039
>. So is the frequency changing? Or is it PWM type sheet?

You have to be joking. How can you be familiar with terms like PWM and be so clueless about a simple motor diagram?

>>1592037
>phase shift
I gotta read up more on this. But that’s what I was kinda talking about, like does “H” chop it up into 3-phase so it’s kinda like a higher Hz?

>>1592041
I’m saying if 1-2-3 in that diagram are all 115V, what is the difference between the electrons coming from 1 and 3 that cause the difference in speed?

>>1592043
>what is the difference between the electrons coming from 1 and 3 that cause the difference in speed?

electrons. lol.

look at the diagram. 1 2 and 3 are tapped into the winding at different places. this makes it like a motor with 3 different windings. the different windings are different and therefore the speed is different.

>That pic doesn’t help.
If the pic won't help, an explanation won't help either.
You're just clueless.

>>1592042
This comes close to your fan motor but it only has two speeds. The 3µF capacitor is the running (or split) cap that shifts the phase of the auxiliary winding to create a rotating magnetic field. There's only one frequency, 60Hz.
Image lifted from
http://what-when-how.com/electric-motors/single-phase-induction-motors-electric-motor/
Scroll down if you must, but the description on this page is worth a read.

>inverter is basically a DC psu of some sort (PV panel, for example) and a H-bridge powering a transformer
>as the h-bridge switches, the transformer generates AC

Now... doesn't that mean a grid-tie inverter can be back-powered the grid when the DC source isn't producing anything?

>>1592067
they could if they were designed to do so, but the transformer doesn't work the way you think. if one is present it's a high frequency transformer used to generate a mains level dc voltage that's then switched with the bridge and put through an LC filter of some sort to generate the mains. a 50/60hz transformer would be too large.

i remember dave jones doing a teardown of a UPS that had an elegant bidirectional converter if you want to go hunting for that.

I hope this is the right thread to ask. I've had this old microphone laying around that my dad found and gave me a few years ago. I've always wanted to convert it to use with my pc. I have no experience in electronics, so it's just a pipedream right now. Maybe I can get some advice here. I took a picture of the inside and bottom of it. It has a spot for 9v battery, but ideally I'd like to have it get it's power from the pc. Wouldn't bother me if it had to be two cords (3.5mm jack and usb) but obviously if it could do it all through one connection that'd be great. On top of all that, I'd like to keep using the push-to-talk button that it has built in.

>>1592075
How do I learn how transformers work? Tesla used to just wind them up for fun

>>1592079
try
https://en.wikibooks.org/wiki/Electronics/Transformer_Design

>>1592079

/ohm/ is having a bad day, I guess.

>>1592077
>It has a spot for 9v battery, but ideally I'd like to have it get it's power from the pc.
>>1592077
>I'd like to keep using the push-to-talk button that it has built in.
The amp inside only uses the 9V battery when the button is pushed.
A battery normally lasts months to years in normal CB/Ham use.
Leave in unmodified to sell on eBay when you get tired of playing with it.

>>1592079
it's pretty simple from an engineering perspective. here's the formula for the number of primary (input) winding turns:
n=V/(cBAf)
>c = waveform coefficient, 4 for square and 3.33 for sine i think
>B = flux density limit in teslas, which you can think of as a material property of the transformer core
>A = cross sectional area of the core that the windings are wrapped around
>f = frequency of the applied AC

the output voltage is determined by the ratio of secondary turn count to the primary turn count. that's about it if you just need one input and one output. everything else including the wire gauge, core size, and core material is just to improve magnetic coupling between the windings and to make sure the transformer doesn't get too hot.

>>1592088
Oh I didn't know that. I guess that would be alright then lol. Any advice for wiring it for the pc?

>>1592053
I was only asking somebody to spoonfeed me!

>>1592091
no, your photo conceals all relevant information like model number and wiring

What could be the yellow part? I guess it's some kind of diode from the marking. The marking is:
106J
M0700

That's a sot23 above for size comparison.

>>1592144
>>1592144
My guess from the base is this model.

shielded white wire is audio
shield goes to ground.
red and black are switch contacts wired to suit radio being connected to
If I recall correctly, the switch on the bottom bypasses the internal amp negating the need for the battery.

>>1592155
tantal cap

>>1592160
thanks

>>1591593
I use Microchip PICs for embedded designs. They have MPLABX that runs on Linux, which is my setup.

I have no clue how deep you want to go down the rabbit hole, but the PIC32 line is 32 bit and has a family of chips designed with tons of PWM and related peripherals just for motor control.
They offer free C compilers for their compilers, as I'm assuming that is what you'll be comfortable coding with. But if you are into it, you can code in assembly for ultimate control.

>>1592164
>They offer free C compilers for their compilers
Since when? Wasn't it always Microchip's thing to make you pay for the C compiler?

>>1591943
Disconnect the resistor and measure it by itself, and also remeasure the digi pot without the 4.7K attached.

Can I just hook a 15v PSU up to a 12v lead-acid battery and have it charge? Or does a battery charger vary the voltage during the charge?

>>1592156
Yeah that's the model

>>1591970
Power supplies in series just add the two voltages, and the3y can be different. For example a 1.5V AA battery and regular 9V batter. In series you'll get 10.5V

In Parallel, unless the voltages are EXACTLY the same, the voltage sources will fight each other. Using small batteries, you'll see interesting results, usually a voltage that is pretty middle ground between them.

A computer PSU in parallel uses PWM topology, and the voltages out will be slightly different. Think 12.1V vs 11.9V. The PSUs will fight each other for a stable voltage, and tend to shut off if they detect something like that. You can pull a LOT of amps from one of those.

As for cutting the green wire, an ATX PSU turns on when you connect the green wire to ground. If you cut it and leave it disconnected, it will just shut down.

>>1592164
>you can code in assembly for ultimate control.
but you sure wouldn't want to
also don't forget to mention the terrible condition of in-circuit programming for PIC and having to buy ever newer programming adapters to keep up with newer chips
whereas Atmel's ISP protocol is pretty standard across their whole line and requires only an SPI interface

>>1592178
no. 2.5V/cell is too much, you'll just gas off hydrogen and damage the capacity. google "lead acid charging profile"

>>1592167
Years ago, yes. Back in the days of MPLab 8 and before.

They released the MPLabX series that is mutli-platform and introduced free compilers for all architectures.
Of course the free version has no efficiency options, and you can still buy a "pro" version that can get your compiled code more compact.
Still haven't had any issues with the free ones though. If I ever needed more space I would buy the PIC with more memory for $0.20 more.

https://www.microchip.com/mplab/compilers

>>1592184
>no. 2.5V/cell is too much, you'll just gas off hydrogen and damage the capacity. google "lead acid charging profile"
What if I find a power supply that can give 14.1v?

>>1592195
>What if I find a power supply that can give 14.1v?
It would be ok but you'd have to monitor it and disconnect it when the battery was fully charged.
At 13.5 - 13.8 volts, you could pretty much leave it for extended periods.

>>1592184
No, assembly on PIC32's are insane, no arguing there.
>don't forget to mention the terrible condition of in-circuit programming for PIC and having to buy ever newer programming adapters to keep up with newer chips
No? Their ICSP uses specific lines for programming, most chips with multiple options and work perfectly. You also don't need adapters for newer chips. They released the $15 "SNAP" programmer/debugger and is widely compatible. The PICkit is a female DIP style connector. Until you step into their ICD3/4 system do you need the $10 adapter to connect to a breadboard (Which is just an RJ-11 plug).
Yes, you can go up the tool chain to their $500 Real ICE unit that can debug real time, let you pause the chip real time, read/modify registers, unpause the PIC and let it keep going. If you need that power. It has a lot of pricey adapters.

>Atmel's ISP protocol is pretty standard across their whole line and requires only an SPI interface
Nothing wrong with that.

Microchip has aged very well. 10+ years ago I totally agree with you, they were rough around the edges. I've been using PICs since before they offered ICSP. Its now 2019. Things changed a lot.

And just for the record, Microchip owns Atmel.

>>1592194
meh, PIC has slightly better on-board peripherals but the programmer model is low-tier

>>1592199
but still shouldn't as you might be wearing the battery down severely without a definite current limiter
by which point it's probably cheaper to just get a battery charger

>>1592203
>without a definite current limiter
auto alternator - 250A
You're book smart but no idea of how real life works.

>>1592203
>PIC has slightly better on-board peripherals but the programmer model is low-tier
I think they have 4 different programmer options, from $15-$500, depending on what you need.

Programming is programming, it just takes more time with cheaper ones. I don't mind saving $100's if I have to wait 10 more seconds.
I like the debugging power of the PICs. You can literally pause the PIC, browse the live code/registers, modify registers, and resume your code. Its saved me a lot of time and hassle over the years.

>>1592206
Not every lead-acid battery is a car starter battery and designed for such large currents, Mr. real life.

>>1592206
>250A
sounds like one hell of a beefy truck you're running
anyway auto batteries are made for that sort of abuse. I don't think I'd want to try that with a gell cell

>>1591082
>Can we lynch him? (bbc asking)

>>1592210
>Not every lead-acid battery is a car starter battery and designed for such large currents
Not every 14.1v power supply has the current capacity to burn up lead/acid batteries.
He didn't specify the battery or power supply.
It's no skin off your nose, let him try it.

>>1592212
>sounds like one hell of a beefy truck you're running

>>1592222
>factory 100-105A
>needs 200A more for their boomin' system

>>1592224
Just driving by this thread, but a kilowatt for a speaker system and a kilowatt of head room to not burn out the alternator seems reasonable to me.

>>1592167

It's kind of hidden and certainly not obvious, but they have links to download their compilers for free including the 32bit compiler. It does require you to make an account, but they do not ask for payment info of any kind to receive the compilers.

Source: I use their compilers for free.

>>1592178
>12v lead-acid battery
suddenly becomes:
>>1592212
>a gell cell
Just make shit up to suit your narrative.

>>1592231
>implying there are no 12V gell-cells
OP wasn't exactly specific what kind of lead-acid battery he was using. stop trolling

>>1592203
>>1592208
segger jlink debuggers are $20 for non-commercial use and cortex-M chips use jtag. My personal preference now is STM32 with stm32cubemx configurator and gcc/gdb in vscode but if you want a full IDE segger embedded studio is free aswell for non-commercial.

>>1592155
Thermite incinerator bomb if you solder it with reverse polarity

>>1592263
Good setup, STM32's are powerful
Any reason its all non-commercial?

>>1592290
segger embedded studio for commercial costs upwards of 1k+. same with the jtag debuggers even though they are hardware wise the same, $500 vs $20.

Technically you could use it for commercial though as they have no way of knowing.

>>1591072
Any high speed shift register out there? I need something 8 or 16 bit that can tolerate inputs of about 50 or 100 MHz.

>>1592313
Try 74VHC164 (SIPO) or 74VHC165 (PISO). They should do 100+MHz.

>>1592199
I hate lead acids. The CC/CV Li-Ion charging is simple to me even though the consensus seems to be that charging Li-Ions is dangerous. But then lead acids, I heard somebody say to charge at a higher voltage to get rid of sulfation and there’s no basic C-rating for them like 18650s so I just put the damn power supply at 13.8V and a couple amps and let it go until the battery is reading high 12s steady. Plus then you disconnect it when it reads 13.2 and half an hour later it settles down at 12.7 or some shit.

I guess it doesn’t really help that I’m always playing with half dead lead acids.

Is it worth springing for a better solder station than the cheap $30 kit I already have, or would it be more worthwhile to practice my skills on spare PCBs?

>>1592357
Try replacing the tip with a chisel tip first, that's a lot easier to work with.

>>1592359
Will do. I was having trouble with the solder blob always staying on the side of the tip, so I hope that'll help.

>>1592154
>he hasn't heard of thermal runaway
You always need some form of current-limiting with LEDs because they have a negative temperature coefficient. If it heats up by one degree, the effective forward resistance of all the diodes will drop, meaning they'll draw more current, and heat up more, etc.

what is a sechspulsbrückenschaltung

why does everyone laugh at the word in greatscott's videos

>>1592263
programmer model as in the processor architecture. a W register is so 1970s

>>1592208
see above. the ICSP situation is a whole other pile of sick. couldn't just stick with a single simple SPI protocol like Atmel did, no, you gotta buy a new programmer for new chips. ftn

>>1592400
six-pulse bridge circuit, according to google translate
>why does everyone laugh
I too was eight years old once

>>1592377
where's your copper wool tip cleaning sponge?

>>1592403
is it a german pun? I only see german speakers laughing at it

>>1592408
I actually ended up buying one of those recently because it didn't come with the kit. I was previously only using the sponge.

>>1592391
Interesting, I'll have to check that out, thanks for the info

>>1592403
>couldn't just stick with a single simple SPI protocol like Atmel did, no, you gotta buy a new programmer for new chips

You don't have to buy a new programmer for new chips, you can literally use a programmer/debugger that came out 15 years ago for even the latest PICs.

They chose not to stick to a "simple" SPI interface because of speed constrictions. Modern day SPI ports can indeed handle high data transfers, but years back when SPI peripherals were slower, they needed better throughput for live debugging.
Hell, even serial communication isn't good enough for a live trace on C coding, you need a whole 8 bit port for parallel data shifting.

>>1592043
Hint: It's not the same using 100% of the windings vs 50% and 25%

>>1591189
I'd recommend getting a used HP 54600A on eBay. You can often find them with the common screen issue that's caused by a bad capacitor. I got mine for $100 + shipping because of that. It's new enough that it basically has all the modern features you'd need but old enough that they can be had for cheap.
>100MHz bandwidth
>digital measurement functions and autoscale
>54658A expansion module adds memory, RS-232 communication and additional math functions (including FFT)
>excellent UI layout and HP build quality
>Tetris easter egg

>>1592359
Or knife tip.

>>1592290
Atollic TrueSTUDIO is also pretty nice for a full IDE, free as in free beer if you're using it on an STM32 including for commercial use.

>>1592490
This is what I was thinking, but they are all coiled into one big bunch and nothing separates them. If 3 copper wires are twisted together and you energize the end of one of them, wouldn’t all 3 of them become energized?

I’m sure it’s something simple, please be patient as I rode the short bus to school.

>>1592532
It's acting as an autotransformer. Apply 115V on blue and you'll get 115V across the whole winding, apply 115V on red instead and it'll boost the voltage up.

If you short 1 2 and 3 together and apply power you'll get windings burning.

>>1592534
>1 2 and 3 together
control board doesn't offer that option

>>1592160
>>1592269
Both of these are correct.

>>1592534
> It's acting as an autotransformer.
That's incorrect. While it's true that applying power to red or white will cause blue to be at an increased voltage, no current flows through that part of the winding. The portion of the winding between blue and red/white could be removed without consequence.

What's actually happening is that the back EMF is proportional to the number of turns, and the rotation speed is determined by the point at which the back EMF equals the applied voltage (minus any voltage drop due to the winding resistance and current). So fewer turns on the winding means the motor spins faster for a given voltage.

I am a bit confused about pull up resistors.

So say you have an input pin which you pull directly to ground with a switch. when the pin is not being actively pulled down you want it to be in HIGH state instead of just floating, so you use a 10k ohm resistor and connect it between that pin and VIN. (10k because that is what pretty much all examples use)
but 10k resistor not only limits current but also drops voltage, so 10k will drop it to like 1V or even less so when the VIN voltage gets to out PIN it won't be high enough for that pin to be read as high by the device (usually to read as high it has to be something like 3V at minimum if the device operates on 5V like an arduino for example).

So how does that work?

>>1592589
Here a simple schematic.
Digital pin is being pulled high from the + rail on the arduino through a 10k pull up resistor, so reading that pin will give a HIGH state.
If the button is pressed the pin is pulled to ground and reading it will give LOW
but like i said, since 10k resistor drops a ton of voltage, then the pin will never read high in the first place, even if the button isn't pressed, since the resistor will drop like 4V a pull the pin to only like 1V

>>1592589
>>1591328 shows an open-collector output. With current being fed into the base of the transistor, current can flow from the collector to the emitter with very low resistance, pulling the output to ground. But when there's no base current and hence no current going through the collector, it needs a pullup else the output won't be at Vcc.

You're right that trying to pull any current from the output pin will result in a significant voltage being lost across the resistor, but for most logical purposes, the current into a device will be in the realm of a few microamps, sometimes even into the nanoamps, making the voltage drop miniscule through V = IR. Usually this is classified in terms of an input impedance, and a higher input impedance (MΩ or sometimes GΩ) is preferred for this reason. The pullup resistor can be approximated as an output impedance, which is preferably made as small as possible, but of course a 1Ω pullup will mean a lot of power wasted. So we make pullup resistors to 10k or so because input impedances are usually far higher than this, but if we want to power a load with significant current, we'd use a push-pull or totem-pole output, something along the lines of pic related. Powering MOSFETs is often done with a totem-pole because we need the gate current to be as high as reasonably possible in order to get the switching times to be small.

>>1592593
Oh i see, so because the current going into the input pin is so tiny, then almost no voltage is dropped across the resistor
that makes sense

>>1592595
I just can't figure out why my circuit isn't working then.
I am trying to trigger an interrupt on arduino by pulling a pin to ground, but it just doesn't work with a 10k pull up resistor between vin and the pin, but if touch the pin directly to vin without the 10k resistor to pull it high, then it works fin.
But that of course sucks because i can't not have a resistor there since pulling it low with a switch would mean the vin would be shorted directly to ground

>doesn't have op info
>doesn't have ohm I'm the op post
>some dumb roastie that isn't even ohm related
why did we choose this thread over the other one?

>>1592597
circuit diagram would help
it's possible that you're mixing 3.3V with 5V, or any number of other pedantic issues

>>1592603
because it has /ohm/ in the title so it can be found by going to >>>diy/ohm

>>1592613
>>>/diy/ohm is what I meant to write

>>1592603
>>1592613
>>1592615
Oh the old one worked because there was /ohm/ written in rule 4, and in the title of the OP image.

But to be honest I'm more upset that we don't have any "this thread brought to you by" IC.

>>1592613
Here is the diagram
>>1592592
It is the most simple circuit in existence and it should work i am definitely not mixing pins I checked like 5 times. The board i am using is the attiny 85 digistump with digital pin 4

>>1592585
The chosen number of turns relates to torque, not speed. Without load torque (here fan torque) all settings lead to the same shaft speed close to the synchronous speed (pic) and the slip is near zero. The synchronous speed only depends on the frequency and the number of poles. This is not a multi-speed motor. The actual fan speed is governed by the torque balance which results in a certain relative difference between the shaft speed and the magnetic field speed, called slip.

>>1592617
Shouldn't you be using 3.3V instead of 5V for an ATtiny85? I'm assuming it's on one of those boards that has a 3.3V regulator on it, but apparently the MCU itself can run on anything from 1.8V to 5.5V, which is neat.

How do I interrupt an HDMI signal? I need something as simple as a switch or on/off button.
Are there any devices that do the job, or do I need to concoct something myself?

>>1592640
buy an hdmi switch and only attach one end maybe?

>>1592640
KVM with HDMI is the low-tech COTS way to do it. Even it just means turning the kvm on/off, or switching the input.

>>1592651
That might work, thanks.

>>1592652
That is also a good idea!

>>1592637
I am bypassing the regulator. But it seems the problem is only with interrupts. With just digital reads it works fine

>>1591072
>>1592616
oh alright here you go.
brought to you by the LT1072 universal switching regulator, and dedicated to that guy from Taco Bell who went to work for Linear Tech, sent me a handful of these things as samples, and is presumably now retired and posting in trap threads on /b/

>>1592782
How do these things work? If you wanted a 5V power output to charge USB stuff, nvm the current, is it as simple as putting ~4v from a Li-ion through one of these things and getting 5V out?

There has to be more to it, right?

>>1592859
no, linear regulators can only drop voltage
you need a boost regulator. and yes, there is more to it

>>1592859
No you need around 7V to get 5V out of one of these.

>>1592864
>>1592866
So those chips would almost be like a very basic buck converter?

>>1592888
No, linear regulators aren't switch-mode devices.
They regulate to a lower voltage and that is where the similarity ends.

>>1592888
How do you know what a buck converter is but not a linear regulator?

>>1592899
Because I’m a simpleton.

>>1592889
Then for the purpose of dropping the voltage, like you wanted to make a throwaway cell phone charger, could you get one of pic related, stick a voltage regulator thingy in there and then hook up the end of a micro USB or Type C cable and stick a cheap 9v cell on there for half a charge?

>>1592910
Yeah but it would only have a efficiency of around 55%.

what is the english word for that kind of wire?

>>1592923
wire wrapping wire?

>>1592923
wait now that i see it, is verowire actually the name of the wire or that of the company?

>>1592923
Enameled wire

>>1592932
no it was Vero Precision Engineering, dissolved in 1996

>>1592923
Magnet wire
>>1592932
verowire is a company name

>>1592859
How would /ohm/ generate -5 and +5v from a 12V power supply?

>>1592939
12:10 transformer with center tap

>>1592939
buck converter for +5V and buck-boost for -5V

>>1592944
>12:10 transformer with center tap
It's a 12V dc power supply so you fried your primary

>>1592939
depends on how close they need to track and how much current I need. for heavy +5V loads and light -5V loads I might use a buck for 5V and a switched-cap inverter like the ICL7660 to create the - rail

>>1592950
No fair, teach. DC wasn't specified in the question.

>>1592939
I'd get two 5V isolated DC-DC supplies off the shelf, because life is too short to design power supplies (at least for me it is)

>>1591072
I have an old 80s stereo that used a wired bulb (stanley 158) that burned out. I cant find any agreeable information on the bulb other than it COULD be 14v 3w and 0.24A

Would testing the power going through the leads be any help? I have this schematic of a similar (not same model) which i believe is the same way on my model. it gets power right from the power amp.

>>1592934
>>1592938
>>1592926
alright thanks. still get mixed results when looking for that stuff. in Germany we call it "Fädeldraht" which means something like "threading wire"

>>1592960
And here is the bulb in question

>>1592960
>it COULD be 14v 3w and 0.24A
what makes you think that

>>1592960
follow the wires in the diagram
lamp -> 5 is ground
lamp -> 6 cinnected to j2->6
ac 12.5v between j2->6 and j2->5
j2->5 is ground
so it takes around 12.5v i would say.

it could be a 14v lamp that is certainly possible, it would run a little dimmer on 12.5v than 14v max.if you are sure its a 158 then there isn't much else to tell you i guess unless someone has the stereo and knows. but you didn't say what stereo? so maybe someone will recognize the schematic?

where did you get the 3w and .24A from?
14v@.24A is 3.36W
12.5v @ .24A is 3W exactly.
a 3.36W 14v lamp at 12v will draw closer to 2.7W than 3W i THINK

>>1592960
>158
>>1592967
That appears to be a standard 158 wedge type lamp that has the regular wire contacts straightened out and wires crimped on.
It could be something the factory did or a specialty item they had made for them
Just go the the auto parts store and buy a 158, straighten out the wire and hold them against the crimps.
If it seems to be correct, make a permanent change using the plastic housing and wires from the burned out one.
(swap out just the lamp)

>>1592974
>>1592976
when i search for 158 equivalents i get this page:
https://www.bulbs.com/product/158
>CEC 3.36W 14V 0.24A Mini T3.25 Bulb

>>1592980
bulb wont explode if its the wrong voltage right?

>>1592976
It's a Technics SA-370 stereo (identical to models SA-937). I guess its just not popular enough to be talked about for replacing bulbs. Its not the end of the world if i dont get a new one, it just backlights the LCD screen showing what radio station youre on.
>pic related

>>1592982
>bulb wont explode if its the wrong voltage right?
right

>>1592992
okay great. do you know what these little sliders are called? might modfiy the wires so the new bulb can slide into these.

>>1593043
why, it's not like you'll likely need to change it again, pretty much ever

>>1593043
I think they call those Faston connectors.

>>1593043
spade connectors?

>>1593043
If you test it and it appears to be working correctly:
slip heat-shrink tubing over the individual insulated wires
strip the ends of the insulated wires
twist the stripped ends with the lamp's straightened wire pins
fold them flat and insulate them with the tubing
if you don't have tubing, use electrical tape
if you don't have electrical tape, you shouldn't be trying this

>>1593061
Oh yeah that makes more sense

>>1593065
don't mess with the lamp pins too much
it's steel wire and will break if flexed too much or bent sharply at the glass bond

reverse engineering smd boards is an effort in futility

>>1593105
Sometimes you wish they'd be a bit lighter with the silkscreen. At least you don't have one of those god-awful black solder masks to have to peer through.

>>1592585
The voltage is still stepped up even when current's not flowing in that part.

>>1593105
hope they're using a reference design, or be willing to destructively strip the parts off the board, ig

Do reed switches need debouncing?

>>1593212
Yes

>>1593212
No

>>1593223
For digital purposes they do. At that point using a hall-effect switch might be more painless.

>>1591161
Not the case i know but forcing ocr in online pages works wonders for me when i have something with text scaned even if its not pdf its faster to convert it in most cases

>>1593236
That's somewhat surprising, but usually scanned datasheets are in good enough condition and with a non-intrusive font. Looking for another datasheet by someone else can sometimes give you a digital version. Weren't captchas being used to train some sort of google OCR AI?

>>1593227
But halls waste energy. Reed uses 0 when open

>>1593266
If you need debouncing I assume you're powering something digital, which will be on all the time anyway and probably using more power.

>>1593267
The ic will be asleep and use micro amps, how much does hall use?

>>1593295
Depends on the hall, but a reed debouncer circuit will require a comparator or some integrated schmitt trigger. If you're using the reed switch as an interrupt and/or to wake the MCU, then you can likely do without any debouncing, but for counting or that sort will require it.

>>1593301
>reed debouncer
something like this

>>1593313
I don't think the second resistor is entirely necessary, but yeah.

>>1593316
Right, it's probably because you normally don't short a capacitor.

>>1593301
>but a reed debouncer circuit will require a comparator or some integrated schmitt trigger. If you're using the reed switch as an interrupt and/or to wake the MCU, then you can likely do without any debouncing, but for counting or that sort will require it.

Depends on what is being counted. We have a reed switch driven by a water meter and it turns on/off every minute or so, so the MCU can easily debounce in software after the first few bounce wakes it.

>>1593323
No it's more to make a ramp on both the up and the down transition, I doubt many circuits will care about even a 1uF cap being shorted to ground, but I might be wrong.
I am now wondering if I needed to use a totem pole for my debouncing circuit.

I have two arduinos.
Each arduino is powered by its own battery pack.

The arduinos are connected by 3pins:
GROUND to GROUND
and 2 I2C pins to 2 I2C

It should work fine, but what happens if for example you remove one battery pack from arduino A, or if it runs out of power?
Won't arduino A try to dry power from the battery pack on Arduino B through the I2C pins thus wasting the precious precious life juice intended solely for Arduino B, or even worse because it tries to power itself on through data pins it will burn both arduinos?

>>1593381
here is a nice simple schema, just a standard i2c

>>1593394
also the sda and scl can only pull the lines low, never high

>>1593334
whynotboth.jpg
reed switches handle very little current. to discharge a cap directly through them would entail high instantaneous currents and therefore be Bad. 47 ohms or so would be plenty in that case

>>1593381
>remove one battery pack from arduino
highly dependent on the particulars of the circuit. any of your scenarios are possible

>>1591652
wow cap company gets 5* for design

>>1592782
>not posting datasheet screen
i gotcha

>>1593586
Man that really hit me in the feels.

>>1593586
based
thanks again, 4chan captcha faggotry

>>1591189
I agree somewhat with HP dude. I would say for a first-scope under the $150 range go for used CRT scope from HP, tektronix, or a few others. They are built to last, no fiddling with menus (and no software bugs), and (subjective) they are really pretty/enjoyable to look at. But for that price, Rigol or Siglent are probably a pretty safe bet. No idea on that trial period but never heard of it; sounds bogus

>>1593105
Consider watching a video of that same name from EEVblog. It he uses some hi-tech equipment, but I remember thinking it was interesting. Might make things less futile

>>1593452
>highly dependent on the particulars of the circuit.
What particulars? it is exactly this >>1593394
where the two sources are batteries

on a phone battery when you see 4 terminals but they aren't each individually labeled (see pic) would that mean that the first pair is negative and the second pair is positive?

>>1593686
>first pair is negative and the second pair is positive?
No.

>>1593394
Why not remove one battery and run their Vcc leads together?

>>1593664
my bad, master will probably just not power up and slave's I2C state engine could get into a strange place. just bitbang a stop condition whenever A powers up and you'll be fine

>>1592939

>>1593764
I forgot to connect C5 to the 7905

Any motor repairmen on here? I have a 1.5 hp, 115/230v motor that's missing a run capacitor. It runs without the cap, but draws 15 amps at 115V with no load and 17 amps with the belt attached. It's an old motor and I'm getting the runaround from motor repair shops and the OEM about the correct uF capacitor for it. Can I just buy a few caps from the 7 - 15 - 50 uF range and see which it draws the lowest current with? Or check the voltage across the cap, and whichever is closest to 115v is the right one? Thanks I don't understand electrics.

I have 220V LED bulb. I want to power it from 36V battery of e-bike.
(it has pretty narrow beam, so don't worry about blinding).
How can I modify its buck converter to boost converter? How efficient is Joule thief?

>>1593807
you're not getting 220V from a joule thief, certainly not at any great power
>modify
you don't. you get or build a boost converter that will provide constant current to the LED string at the appropriate voltage (which you will need to determine) and you remove or disconnect the buck that's already there

>>1593810
>you don't. you get or build a boost converter that will provide constant current to the LED string at the appropriate voltage (which you will need to determine) and you remove or disconnect the buck that's already there
I'm broke. It is easier to make two separate strings and some liner CC thing (cheaper lamps have it) or simple resistor.
Or maybe stock buck converter, but that would run definitely out of specs.

>>1593817
>It is easier to make two separate strings
this is permissible. so how long are the strings?

>>1593819
8 LEDs.
Well, I tried shorting 5 LEDs, stock buck converter limited current without any problems.
I need to try it @32V though. And yeah, frequency became lower, I can hear coil whine.

>>1593820
I guess modern LED bulbs have 3 chips in series in one LED...

>>1593820
>8 LEDs
x 3.7V < 36V. you can use a stock buck converter for this as long as it accepts more than 36V on the input. Pic related is for a boost converter but the principle is exactly the same for the buck

>>1593822
in that case, you're gonna need to do some board hacking and put some strings in parallel

>>1593823
I think modern LED bulbs have 2-3 chips in one LED. So it is around 30Vd and 4 is around 40Vd, so it is too much.
>as long as it accepts more than 36V on the input.
https://www.datasheet4u.com/datasheet-pdf/JoulWatt/JW1780/pdf.php?id=1283323
It accepts IRL, but I think it wasn't designed for such application.
For better oomh I guess I can remove diode bridge, since I have DC on e-bike.
>>1593827
Sure. I need to cut three groves, and get second bulb for another DC-DC. How they manage to make them that cheap?

>>1593830
Oh wait, it needs only 5.5V, so I guess it will be fine at low voltage.

>>1593830
what part of "you can't turn a buck converter into a boost converter" do you not get?

>>1593833
I don't need a boost, since I can cut traces and get another bulb in order to get another DC-DC, so I can power all LEDs.
Driver/fuck-converter will work at 5V, so it is not an issue. (36>5).

>>1593834
So, which color should I get? 2700k? 5000K? or 7000K?

>>1593807
Is replacing your bulb with a 12V COB not an option? They can be pretty cheap, sub $5 range.
Depending on your LED bulb's circuit, HVDC may or may not work. If it's a capacitive dropper, it won't take kindly to having DC across it at all, though potentially you could just feed it higher frequency 36V square waves until it powers the LEDs with a decent brightness. If it's a switcher, you're basically feeding a buck converter with a boost converter, so removing that extra step and just using a constant-current converter (linear or switching, depending on the voltage difference) would be preferable, but not required.

To deconstruct the LEDs inside the bulb will likely be fairly destructive, but if you can (and if I'm reading correctly that you have 16 individual ~3V LEDs that you are able to fuck about with) then wiring them as two strings of 8 LEDs, you'll need to drop up to 12V in your current regulator. Linear is basically out of the question, unless you want to wire up a few more LEDs to make the voltage closer to 36V and ditch the remaining few.
So the question follows, what are you willing to buy, and what do you already have, both in terms of LEDs and buck converters?

>>1593835
>which color
If you don't have them yet, why not just buy a DC LED? Also 5000K.

>>1593837
>If you don't have them yet, why not just buy a DC LED?
I have, but I will be getting similar bulb for LED driver, so why not select proper color.
>Also 5000K.
Absolutely agree. This is exactly what I have.
>Is replacing your bulb with a 12V COB not an option? They can be pretty cheap, sub $5 range.
It might be an option, but aliexpress is slow, and local shops have insane prices.
>Depending on your LED bulb's circuit, HVDC may or may not work. If it's a capacitive dropper, it won't take kindly to having DC across it at all, though potentially you could just feed it higher frequency 36V square waves until it powers the LEDs with a decent brightness. If it's a switcher, you're basically feeding a buck converter with a boost converter, so removing that extra step and just using a constant-current converter (linear or switching, depending on the voltage difference) would be preferable, but not required.
It is switch mode non insulated DC-DC, and I tried it, it worked at 38V. IC will work at any voltage above 5,5V according to datasheet.
>To deconstruct the LEDs inside the bulb will likely be fairly destructive
On cheap bulbs you can remove glass really easily. Getting power supply out is tricky though.
>(and if I'm reading correctly that you have 16 individual ~3V LEDs that you are able to fuck about with) then wiring them as two strings of 8 LEDs, you'll need to drop up to 12V in your current regulator.
Pic related
>So the question follows, what are you willing to buy, and what do you already have, both in terms of LEDs and buck converters?
I have a bulb, and I would need another DC-DC from similar bulb, since there is not enough voltage to drive more than 3 LEDs in series.

>>1593845
So you have 8x3x3V worth of LEDs, or 72V if all in series? And you want to keep them on the bulb's PCB or do you want to desolder them?

>>1593847
>or 72V if all in series?
Somewhere around that.
>And you want to keep them on the bulb's PCB or do you want to desolder them?
Desoldering anything from those alu PCB is pain, heats get sucked immediately, so I will keep PCB, just add couple bridges.

>>1593849
So you plan on using LED bulbs' converters, running them off 36V instead of 340V (rectified mains), and feeding seperate strings from these?
Sounds like it will be more than a couple of bridges, but good luck, sounds feasible. Make sure your trace carving doesn't bridge copper to the aluminium-backing, I hear the insulating layer is pretty thin.

>>1591286
Rolling.

I made a "digital-analog converter" out of a R2R ladder. is there a way to smooth the output out? What would happen if I used an opamp integrator on this?

>>1593807
Sounds like a visitor from /sci/

>>1593861
I never visited /sci/ though.
>>1593854
Yep.
I meant cuts, not bridges though... But yeah, insulation is thin, so I will use dremel.

>>1593859
Wouldn't an integrator turn it into a delta-sigma ADC? Either way, wouldn't a normal low-pass work fine?
Not sure if the ladder design would be any more precise than PWM and a low-pass or integrator anyhow.

>>1593859
Anything that cuts out frequencies at or above half your sample rate. Integrator can work but the usual way is just a low pass.

>>1591286
Roll

How's your PPBUS?

>>1593728
Because master will be connected only some times, most of the time it will be just slave alone doing slave things.
Plus i will have multpiple slaves and each needs to have its own battery

>>1593731
there must be a better way since i need to hot swap slaves and masters in the system (there will ever only be one master at a time tho).
I was thinking optocouples to completely separate them but that seems needlessly complicated

>>1593926
Well fine, but if you have wires between them at all times anyway it would be better to just use one battery. I'm actually curious what you're doing with these, is the master a hub for a bunch of sensors that are at different positions and recording data that only syncs to the master once every so often?

>>1593939
Well basically you have a bunch of energy meters, like water and shit.
Each meter has a tiny slave chip on it which counts energy pulses given by the meter, can be magnetic or ir whatever. The slaves can be interconnected via i2c but don't have to be they can be completely isolated too.
But at any time i want to be able to simply plug the slaves into a master's i2c line and read them (for example hand held arduino master you walk around with and plug it into slaves to read their data). If the master's battery dies or the master gets completely disconnected is fine. But the slave absolutely cannot at any time run out of batteries it must be up and counting at all times,, no mater what other devices are on the i2c line

>>1593967
Then what's wrong with having the slave's battery power the master when the master is plugged in? The opposite, having the master's own battery help charge up the slave's battery is also a possible solution.

>>1593968
Slaves are tiny little boys with tiny little battery penises, the master nice a large bbc to slurp it's power from. If it ever tries to operate from slave power it will drain weeks if not months of slave operation time on that battery.
Plus i want to use audio jack for the master connection to slave network which has room for just gnd, scl and sda
there must be a way to make this work without the batteries on different dvices fucking around with other devices on the network

>>1593969
There are TRRs jacks, but I can see where you're coming from. I don't think you'll have any problem with the batteries, but if you do then I have an idea or two. Putting a FET on each line like a level shifter might do the trick, so the master side will never go higher than the master's Vcc, but that does require both voltage rails to be present at the same place, i.e. a 4-pin jack. If you do go this way (and it works in simulations), then using the master's power source to recharge the slave batteries would be a nice convenience feature.
Have you considered using optocouplers?

>>1593974
>optocouplers
I did but only as a last resort since it seems overly complicated if a simpler solution may be possible.

The 4 contact jack might not be a bad idea but i need to make sure with fets or something the i2c lines are blocked until the jack is plugged in because the data contact can't touch before the power contacts

>>1593980
I don't think the contact order would be an issue if the FETs were on the slave side of things. But the FET method is arguably more complicated than using optocouplers.

>>1593982
What would be the easiest octocoupler solution here?

Don't forget that both master and slave have to be able to pull both lines low if the want to.
If it was only master talking it would be super easy, just two octocouplers, but if the slave want's to reply it will be a bit more complicated

>>1593773
https://www.quora.com/How-do-I-calculate-capacitor-value-for-single-phase-motor

why are transistors used to control LEDs?

>>1594002
because it can be done

>>1593987
Oh yeah that would be a pain, using the FETs would be a better idea then. Especially with a dual FET package, perhaps crammed in a SOT-23-5/other 5-pin package with common gate, assuming those exist for this sort of esoteric situation (they probably don't).

>>1591661
thats called a distortion pedal.
eqd makes the acapulco gold, which is suposed to sound like an power amp clipping, but its just two 386´s cascading into eachother.

>>1591671
i don´t like compressors, i want to be loud or more distorted when i play harder and less loud or less distorted when i play softly.

Wow it really works, i had no idea battery backup power was this easy

>>1594074
i hope you have beefy diodes. what are you using this for?

>>1594078
Just ic things

>>1594081
just ic things? there are ic´s capable of handling lots of amps, but good luck to you

>>1594074
Vf of top diode looks fishy to me.

>>1594082
it's only logic level stuff, 200ma at absolute maximum for short bursts

>>1593927
add very large (like 1Mohm) pullups on the slave side, then. if your micro has CMOS inputs, the drain should be very close to zero. thus alleviating any potential for bad I2C states as long as the bus is idle when unplugged

>>1593974
>optocouplers
>I2C
stop posting

>>1594103
the BAT54A is your friend

>>1594002
what else are you going to control them with?
the GPIO pins on your tarduino also use transistors.

>>1594074
>i had no idea battery backup power was this easy

your simulation is whacked out. there's no reason for the top diode to be dropping 1.25V.
consequently, the current from the battery should be near zero as the lower diode is substantially completely off. to make double-plus-sure it's completely off, use a Schottky diode at the top and a silicon one at the bottom.

>>1594205
plausible for a 1N4148 being run well beyond its ratings

Microphone guy here
>>1592077
So I spliced an old 3.5mm jack I had lying around to the white shielded wire and the microphone worked for a short time. When I put it back together, it stopped working so I took it apart again and found that the little yellow wire on the inside had broken off, so I resoldered that sucker back where its supposed to go but now I'm still not getting anything. Any advice?

Is creating a makefile going to get significantly harder when I add i2c sensors?

>>1594289
not if you set it up right in the first place. you might have to create some #include files with prototypes if you're feeling especially lazy, which can be a little annoying
a simple, incomplete Makefile from one of my older projects:

CC=avr-gcc
CFLAGS=-ggdb -mmcu=atmega16 -O2
TEST_OBJS=test.o uart.o

test: $(TEST_OBJS)
$(CC) $(LDFLAGS) -o test $(TEST_OBJS) $(LIBS)

.c.o:
$(CC) $(CFLAGS) -c -o $@ $<

>>1594131
>stop posting
xd

>>1594307
build and post webm

>>1594287
okay, I think I've got it. For some reason it won't work with the ptt on the microphone, but it works if I have the lever hold the ptt down all the time. Maybe it's something software related? Any ideas?

>>1594317
the PTT switch disconnects the mic. short across it

>>1594321
elaborate?

best LCR meter? Is this it?

>>1594317
>Maybe it's something software related? Any ideas?
What software does this microphone run?

Retarded junior EE student here, shouldn't this be 0.1e^20t ? Is the Y just a typo or am I missing something.

How sketchy would it be to use Chinese electrolytic capacitors off of aliexpress for high voltage stuff (oscilloscope power supply)? I'm conflicted because on one hand, some of these parts are complete mystery meat made by brands you never heard of, but on the other hand, a lot of cheapo new electronics use them and they don't fail.

>>1594376
"t" makes more sense, but it technically could be either since I'm not familiar with either dy/dt = a*exp(b*y) or dy/dt = a*exp(b*t) with regards to the lorentz force. Shouldn't it be a constant acceleration (linear velocity function) and not an exponential since F = Id x B?

>>1594430
I bought ones from ali and they came with Nippon Chemicon written on them and measured the right capacitance and were a reasonable size without feeling hollow, so you can probably get decent ones. Just so long as you keep in mind that if it sounds too good to be true, it probably is.

>>1594334
>$400
jesus it better be

>>1594430
I'd order them factory-fresh from LCSC or Digi-Key if possible, unless you can verify the manufacturer and the series, and that it is appropriate for the application

>>1594440
>I'd order them factory-fresh from LCSC
How did I never know of this site before? It was in the sticky this whole time and I never noticed it. Although the prices for components from all the big names seem to be average, they have a cheap economy shipping option. I swear, sometimes when I ordered stuff off of mouser shipping would make up half of the cost, and that was when I already had a list of a bunch of components.

>>1594462
they also have plenty of low to medium-tier Asian brands that appear to be good enough for consumer electronics. I see Chengxing 400V caps all the time on boards on Ali and no explosions, so they might be okay
>mouser
their flat-rate shipping is why I started using Digi-Key domestically. they bill actual shipping charges and will use anything down to USPS first-class if your order is small enough
>LCSC
I'm waiting for an economy package from them right now. they and their sister company JLCPCB use 4PX which uses DHL e-Commerce on the backend, which isn't too bad, about 3 weeks to get my order

I found this time another power supply in junk. This time from HP laptop, fully working, except shitty cable...
How to stop hoarding shit from recycler?

>>1594469
turn around 180° instead of 360°

>>1594470
That wouldn't work, since I don't go to that place specifically, but just pass it.
Simple recycling bin in uni... And I'm not alone, sadly. Someone stole fully functional Nokia 1100 with badass LCD.

>>1594474
>>1594469
Anyway, how do I disassemble those fucking supplies without trashing case? o

>>1594478
you don't, they're glued or welded together
you have a problem

>>1594480
I know, but repair niggers manage to disassemble them without any cracks... They won't tell secret even for beers. Put dick in the vice?

>>1594482
gasoline. did you even fucking google
https://www.youtube.com/watch?v=a00esZ3B-aQ

>>1594483
I did, but not in English. Based pajeets.
But won't it eat though plastic?
What other solvents might work, since sucking benzine from car with anti sucking technology sucks. Kerosene? White spirit? Acetone?

>>1594150
Well... I don't know, that's why I'm asking

Usually I just stick the LED in series with a resistor with the rest of the circuit

and usually I see people saying "Control the LEDs with a power MOSFET" when they mention driving a group of LEDs
and I just don't have the experience to know/say why

>>1594484
get on it and report back

>>1594489
Acetone (part of silicone grease spray) worked well.
Now I can be rich of this.

Do I need this?
My brain says yes, but my wallet is crying in the corner again.
https://www.youtube.com/watch?v=il86j54dTzc

Hey, anybody can recommend a chink soldering station? A hakko clone maybe?
If not I'll buy some random one and post results. (like a month later)

>>1594522
looks chunky

>>1594525
I got a KSGER T-12 station with internal power supply, it's pretty good value and I haven't had an issue yet.

>>1594525
ts100 is all you need son

>>1594528
god no

>>1594569
oh no! it's retarded...
i am so sorry boy...

>>1594614
>paying $50 and having to buy an external power supply for the same capabilities of a $35 station
I don't know what to say.

>>1594616
>why buy a $1000 computer when $50 ones exist

I am using an AVR 8535mega to generate a DC voltage by using timer 0 which is 8 bit to generate a pwm that I feed into a resistor and capacitor but it's not where I want it to be on the chip so I would like to change over to timer 1 but it is 16 bits.

I can post the code if need.

>>1594525
Idk, I had cxg mains iron with hakko tips, was good enough.
Never had hot air though

>>1594666
>it's not where I want it to be on the chip

how is it that 18650 power banks can put out 5v when each battery puts out 3.4? do they wire some in series or something?

I want to make my own large power bank for a rpi in the field but need to get a steady 5v supply

>>1594738
Yes. They go in series and also parallel, depending on how many batteries you got. You also need a battery management module in order to output 5v, and to be able to charge the bank without blowing it up.

>>1594738
Every power bank I used had all 18650 cells in parallel and a regulated step-up converter to 5V. This is the simplest structure that needs no balancing.

so say I wanted to join 50 18650s in parallel for a power supply, would I need a regulator board meant specifically for 50 18650s or is it something more flexible

>>1594528
stop memeing

>>1594738
boost converter

>>1594820
you'd simply put them in parallel
you'd better make sure the batteries are all at the same state of charge and voltage before connecting them together or you may be unpleasantly surprised. also you're going to need some custom charging circuitry to handle 50P1S because most Li+ charges aren't going to deliver 50W or so and most have a safety timeout built-in

>>1594628
>for the same capabilities
>FOR THE SAME CAPABILITIES

new thread >>1594926
new thread >>1594926
new thread >>1594926

>>1594927
>not even on page 10 yet
seriously, fucking stop

>>1594931
>seriously, fucking stop

get a life nerd

>>1594970
>>>/g/
>>>/s4s/

>>1594931
>>1594972

I don't understand why it triggers you so much. If it's an /ohm/ rule, it should be stated in the OP, and the OP should not say "old thread reached bump limit", it should say "old thread has reached page 10".

And I apologize for ruining your day. Just ignore that thread, and when this one gets to page 10, make one properly.

JLCPCB 2-layer board order report: bretty gud, economy shipping isn't miserably slow, electrically correct, mediocre silk-screen registration (unsurprising since they're using giant inkjet printers on prototypes). 8.5/10 would try their 4-layer service
/blog

>>1594989
It's an universal slow board thing.
The few fast threads would clutter up the entire catalog with dead threads if they were made at bump limit.

>>1595097
i'm cheap college fag but is the probably the best solution for small(<5 qty) boards? I guess it's a time vs money thing but I'm so sick of wiring up my circuits. I have to spend what seems like an order of magnitude longer on the board layout(++ troubleshoot layout) vs basically every other part of the process. I can imagine I'm a tard in this respect because I have trouble with organization of everything but I can imagine this is common frustration.

basically this turned into a rant but how much might it be for a small order of say 10-15 cm 2 layer

>>1595113
cm^2

>>1595113
yes, prototype pcbs are god-tier. absolute must if you're at all serious about electronics
once you get the hang of using layout software, it's easy and even enjoyable
usually board houses don't even care about the exact size of your board as long as it's smaller than 5cm or 10cm (depends on the board house) in any dimension. within that size limit, JLCPCB's regular price is $5 for 10 two-layer boards with reasonable specs. they will do 5s but it's only $1 cheaper. unless you really need to save on shipping, just get 10, you'll find plenty of use for 1.6mm shims around your workbench
that said, if you have any non-standard requests or need it in your hand sooner than 3-4 weeks, you should fill out the quote form at pcbshopper.com and get simultaneous quotes from 20+ board houses with a variety of turnaround times

does a diode always drop .5v regardless of voltage or current?

>>1595137
no

>>1595140
that is almost a straight line just slightly bent, so the drop is sort of consistent then

>>1595142
the y axis is logarithmic. try again
https://en.wikipedia.org/wiki/Shockley_diode_equation

I have an old GPS that my dad insists I try to fix since he loves it or some shit. He thinks it's the battery, since the USB connection went wonky, and would sometimes work and sometimes not, before it completely stopped working. I tried soldering a USB cable directly to the pins, but the whole thing just seems to heat up now, so I know I fucked something up.

The battery is 3.7V lithium, it shows about 4V when off, and 3.65V when turned on for a few seconds before it shuts off. I'm assumming it's fully charged then, and it's not the battery? Is there a way to check?

>>1595415
Can't help much but I can say that the battery is really old (I'm guessing it's many years?) and you might want to consider trying to check current draw. When a battery drops that much in voltage it's either drawing a significant amount of current (you said it gets hot) and there's prob some kind of short, or it's getting really old. I think both, but if you pop a 1ohm shunt in series with battery you can measure the current draw. if it's in single digit amps, it's hardware; head over to new bread tho

>>1595415

THIS THREAD IS ALMOST ON PAGE 10 I HOPE I GET TO MAKE THE NEXT OP I HAVE A LOT of good ideas don't fuck with me I'm serious.

>>1596069
>I HOPE I GET TO MAKE THE NEXT OP I HAVE A LOT of good ideas don't fuck with me I'm serious.

>>1594926