[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Due to resource constraints, /g/ and /tg/ will no longer be archived or available. Other archivers continue to archive these boards.Become a Patron!

/diy/ - Do-It-Yourself

View post   

[ Toggle deleted replies ]
File: 602 KB, 2104x1112, do_this_do_it_now_I_dare_you.jpg [View same] [iqdb] [saucenao] [google] [report]
1526326 No.1526326 [Reply] [Original] [archived.moe]

old thread >>1522826
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 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.

>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:

>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:
NI Multisim
iCircuit for Macs
KiCAD (pcb layout software, v5+ recommended)

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

>Related YouTube channels:

>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.

>> No.1526350

From reading his entry level book, I can say there's nothing wrong with Mims, but there are also many others who make good books.

When it comes to beginner stuff, there's a lot of authors covering the same material.

>> No.1526373
File: 318 KB, 960x960, -916076834-1341961583.jpg [View same] [iqdb] [saucenao] [google] [report]

Why do they call these shits 3 w leds when they have 2.5V forward voltage and 700mA forward current?

>> No.1526425
File: 1.24 MB, 1201x643, dd.png [View same] [iqdb] [saucenao] [google] [report]

I bought a tiny voice changer from goodwill to modify it into an effects pedal

I wanted to replace the power supply with a AA/AAA harness, but lo-and-behold it takes a tiny battery that ostensibly is 12 volt, 23 amps

what the fuck? I've never heard of something like this. Why would a tiny voice changer need something as intense as that? It uses a SOIC LM386, but I don't think they're *that* power hungry

>> No.1526429


maybe that's so the battery lasts a long time. do you have a meter so you can measure the actual current when it's idle and when it's making you sound like a girl?

>> No.1526430

wait that's 23A as in a literal AAAAAAAAAAAAAAAAAAAAAAA-size battery, it's a miracle I don't forget to breathe

>> No.1526455
File: 45 KB, 972x541, facilierwtf.jpg [View same] [iqdb] [saucenao] [google] [report]

23a is the size dumbfuck

>> No.1526457
File: 78 KB, 661x232, Screenshot_2018-12-30_16-54-30.png [View same] [iqdb] [saucenao] [google] [report]


lrn something new every day.

and forget it tomorrow

>> No.1526460

Hello fellow fellows. I`ve made a linear PSU as starting project, the good thing it has a very low drop out, because I unconscionably made a open collector topology, the bad thing is that in order for it to not oscilate it requires thicc output capacitance. But my real problem is the lowest voltage it can go to, which is about ~1v. My question is
>How can I make a current source that reliablies draw a couple mA across a large voltage range without buying a reference current source IC in order to make it go down to zero or near zero V.

>> No.1526509

While building an elenco fm88k I accidentally soldered the second capacitor on the front of the board, when the instructions clearly say this is the ONLY component that you solder facing the back. Did I fuck it up? A positive lead is a positive lead, and a negative lead is a negative lead. I can't imagine why they would have you assemble it that way, though I am pretty ignorant (first soldering project) so I figured I'd ask here. Is there any good reason to mount an electrolytic capacitor on the opposite side of the pcb from every other component? Pls halp /ohm/

>> No.1526511

if everything physically fit and was electrically correct, then fine.

post pics of instructions and your abortion pls

>> No.1526525
File: 7 KB, 204x324, isense.png [View same] [iqdb] [saucenao] [google] [report]

Consider posting your _complete_ circuit diagram, in form of a readable technical drawing. Name all components, dimensioning can come later. I made you an example.

>> No.1526537

Sounds a but low, I wouldn't be surprised if they got up past 3V.

>> No.1526568

those are classes, not really wattage ratings
anyway >>1526159 is right, you need a constant current driver with the mA rating and capable of at least the series voltage of the array

you need a charger circuit and a power sharing circuit (if you want to charge and discharge at the same time). the MCP73871 is a linear charger that handles both, likewise the ETA6003 is a switching charger with both
then you need a switching boost regulator with enable, and an undervoltage detector. almost any of the billions and billions of boost chips you can buy has an external enable. the undervoltage detectors in the heretofore components are usually set only to protect life and property, not the cycle life of the battery, therefore you have to add one, so add one nanopower comparator and a few megohm-range resistors to your BOM
optional: a fuel gauge, of which there are many different types and levels. you can go from a single comparator with a fixed threshold and a lo-battery light oscillator, all the way to an OLED/MCU gauge

actually helpful. he got me started decades ago

no we don't you fucking mormo. go back to >>>/g/

heat shielding is the first reason that comes to mind for me

/ohm/ RULE 2

>> No.1526599
File: 1.09 MB, 2560x1536, 1546218407309-1482684179.jpg [View same] [iqdb] [saucenao] [google] [report]

This is the front of the board. My solder joints then the schematic coming next.

>> No.1526601
File: 939 KB, 2560x1536, 1546218593254-1770917269.jpg [View same] [iqdb] [saucenao] [google] [report]

My first soldier connectipns. Upon closer inspections, it seems I bridged two leads for the IC socket. Hmmm.

>> No.1526607
File: 1.01 MB, 2560x1536, 1546218702662-542010006.jpg [View same] [iqdb] [saucenao] [google] [report]

Last but not least, the schematic. The two capacitors I didn't install were optional. The electrolytic one for more gain I believe, and the ceramic capacitor for better clarity.

>> No.1526619
File: 9 KB, 413x236, osc32a.gif [View same] [iqdb] [saucenao] [google] [report]

Do I need this extra inverter after the oscillator? I'll be using it to clock shift registers. Also I hope I can use the same circuit for a 32.768kHz crystal.

>> No.1526622

It helps reduce loading on the clock crystal and increase stability if you have many devices running off of the clock.
If you're gonna build this circuit use 74HCT inverters. (high current drive, ttl compatable = 5v clocks).

>> No.1526624

It's only there to provide high input impedance and a Schmitt trigger. You'll almost certainly be fine without it, but there's not really any reason not to get an inverter IC with two circuits.

>> No.1526631
File: 206 KB, 1062x1375, 1525545230541.png [View same] [iqdb] [saucenao] [google] [report]

on the LM386, you mean? the back-side mounting probably prevents feedback into the 7088. probably better pull it out and place it on the back if you get a chance and you haven't already

there's a "little logic" chip made just for this, Pic related. not sure why your diagram called for the first one to be a Schmitt, I don't see that starting very reliably

>> No.1526635
File: 1.60 MB, 3264x2448, oh no.jpg [View same] [iqdb] [saucenao] [google] [report]

I'll be using 74HC14s, I assume there's nothing wrong with that? I want to be able to switch between two crystals, but leaving a crystal floating while I switch the other into that circuit isn't a very good idea, and grounding the other one sounds like an even worse idea, so I'm guessing I need to make two seperate oscillator circuits and switch between the two.

Since I only have 2 inverters left for this, I can't decide how to wire it up. Perhaps switch the two on/off alternately by pulling the inverters' inputs to ground and then combining the outputs with a transistor-resistor-logic AND/NAND gate? I have some 2N5109 RF NPNs that should work for this. I could pull the input high and use an RTL NOR instead, but either way I'd be using a 2-transistor gate. Whether I should use a transistor or switch directly for the first action is also a question worth considering.

I already have all the parts I need, so if the inverter method will work without any issues I'll use it. I'm also using a 74HC04 in the same circuit so I can swap that first one out for a non-Schmitt if I have to, but if I'm swapping between each one by grounding the input of the one not in use it shouldn't prove to be a problem.

>> No.1526656
File: 1.95 MB, 4032x3024, image.jpg [View same] [iqdb] [saucenao] [google] [report]

Hmm... would this be a good starter project for my retarded ass?

That Arduino thing is going to be way too deep. Was thinking about a power supply as well because those seem to be popular, or one of those shitty $25 open source oscilloscope kits.

>> No.1526661

No the 386 is supposed to go on the front, the 220 microfarad cap (c2) says it goes on the rear of the pcb and I can't figure out if that's why it's not working. the only things I seem to have done wrong is that and bridging two joints on the socket for the LM386. The LED lights when it's turned on. No sound from speaker.

>> No.1526676
File: 21 KB, 793x474, 1532320748700.png [View same] [iqdb] [saucenao] [google] [report]

a bog-standard switching transistor makes a fine AND-gate with a wired-OR output. you could also use MOSFETs if you happen to have them on hand. Pic related, at right is a standard TTL inverter as a test load

>> No.1526679

> still avatarfagging
A good starting project would be to go through the expansive tutorials for the shit you bought. We're not here to hold your hand faggot.

>> No.1526686
File: 40 KB, 400x289, 1537893912352.png [View same] [iqdb] [saucenao] [google] [report]

threadly reminder

>> No.1526688

>go through the expansive toutorials
That’s too much work for me! How the hell am I expected to actually finish any of those projects?

>useful 4chan post #73386959768

>> No.1526715
File: 247 KB, 705x527, 1545101248321.jpg [View same] [iqdb] [saucenao] [google] [report]

If the energy stored in a capacitor is equal to the capacitance times the voltage squared, then what's stopping me from making a small capacitor with a really high voltage rating, thus storing a ton of energy in a really small footprint? Doubling the voltage rating quadruples its energy density.

>> No.1526716

let's say I have a 3 pin 10k pot, with a 5v source on the outer two pins
if the wiper is dead center, what voltage is the output of the potentiometer?

>> No.1526722

As it turns out, making a capacitor that stores a higher voltage means you need to increase the distance between the plates; to increase the thickness of the dielectric. The act of doing this means that any solid dielectric of a parallel plate capacitor will store a maximum energy per unit volume that doesn't change as a function of the geometry of the capacitor. Assuming that the plate volume is negligible.

Of course, that didn't stop me from buying 10* 20kV diodes and a fuckton of magnet wire to make a flyback converter to charge up a cubic foot of candle wax + aluminium foil capacitor. Haven't invested in all the fucking wax though. Plan is to do some neat DC electrostatic experiments with it, like ionocraft and electrostatic motors.

>> No.1526725

>Of course, that didn't stop me from buying 10* 20kV diodes and a fuckton of magnet wire to make a flyback converter to charge up a cubic foot of candle wax + aluminium foil capacitor. Haven't invested in all the fucking wax though.
This is why I come to 4chan.

>> No.1526726

Is this also true for supercapacitors? Don't they do some funny stuff with high surface area activated carbon electrodes?

>> No.1526727

Didn't see a stupid questions thread and didn't want to make a whole thread for one question, but thought maybe someone here might know.

I just got an infrared thermometer and I feel like it might be giving me incorrect readings.
Can anyone one think of something I can read that can prove or disprove this?

Also, any tips for measuring hot water temps? I know this isn't the best tool for it, but I should be able to get somewhat close, no?

>> No.1526733


>> No.1526734

They're not true "parallel plate" capacitors, so I'm unsure.

>> No.1526738

What do you mean?

>> No.1526742

Well a parallel plate capacitor refers to that particular geometry. In an electrolytic or super capacitor, the two plates aren't parallel at all on closer inspection, but are a complicated shape with high surface area. It's been a while since I integrated over a surface to find capacitance.

>> No.1526743

>integrated over a surface to find capacitance
How do I do that?

>> No.1526751

Nice, gonna give this a shot.

>> No.1526753

you're getting into some heavy >>>/sci/ now

>> No.1526755

Fuck if I know, it's a surface integral, and without an easy mathematical description of the surface you're not going to be able to integrate it very easily. So I'd go with something fairly simple, like a sinusoid projected into 3D space. Using Gauss's law, Laplace's/Poisson's equation, or something along those lines, I think you integrate all the energy it takes to infinitesimally feed the charge into the plates, at which point C=Q/V gives you your capacitance. But yeah /sci/ will handle that better than I. Hanging around there might bring forth the wealth of wisdom that guy with the electromagnetic simulator software has.

>> No.1526756

This is covered well in either Griffith's Electromagnetics or Serway and Jewett's Physics for Scientists and Engineers. Almost definitely the former but if it's the latter it'll be a lot easier.

>> No.1526758

Haha I remember that first one being mentioned once. Miracle I passed that paper.

>> No.1526759

I can't even get a simple oscillator to work. The led is always on instead of blinking rapidly.
Worst part is that my hands are large and redoing the breadboard is a pain.

>> No.1526765

What topology?

>> No.1526766

It depends on where the wiper pin is connected. If there is little to no load, then it would be 5 V at every position.

>> No.1526773

Update: did some testing all readings are fine and confirmed except hot water in a mug.
Surface temp is much colder than internal.
Obviously that makes sense, just a larger gap than I imagined ~5° - 10° F.

>> No.1526778

Surface of water might be reflecting IR from elsewhere, float a piece of paper atop the water and check its temp instead.

>> No.1526780

Who makes the best datasheets and why is it TI?

>> No.1526781

Because they speak English

>> No.1526788

2 transistors, 2 diodes and 4 resistors

>> No.1526790

I've never gotten a 2-transistor oscillator to work and neither have at least 2 other people on this general, I'd stick to a good old 555 timer, comparator oscillator, or something else more positive.

>> No.1526797

Can I use diode logic at 4MHz with 1N4004s?

>> No.1526801

Probably not, search for "1n4007 reverse recovery time"
Use 1n4148 for a general purpose high speed diode which has a much faster t_rr

>> No.1526807

Looks like I could barely use it at 32kHz. Anyhow it doesn't really matter anymore.

>> No.1526815
File: 41 KB, 577x1024, just.jpg [View same] [iqdb] [saucenao] [google] [report]

>/ohm/ RULE 2
he asked for a current source with a good compliance range. Why the fuck would you need his schematics for this? Go back to the arduino thread retards.

As for the question, you probably have a reference somewhere in your psu, you can it to make a current mirror, even though it`s haram to make one out of discrete transistors it could still work well enough.

>> No.1526817

Look for Homopolar generators for current equivalent of what you are doing kek.

>> No.1526821

I know that feel
needle nose pliers minimize that feel

Analog Devices' datasheets are the worst you can get without street-shitting or dog-eating being involved, afaik

in /ohm/ we pride ourselves on answering the question you needed answered, even if it's not the question you asked. perhaps it's a problem in his feedback loop design that could be fixed with some simple rearrangement

>> No.1526824

Linear PSUs have a minimum load for the pass transitor, you are not /ohm/ you are just a step above beepu if you can`t even sugest a current mirror or a bjt topology for a stableish source. kys

>> No.1526835

You're ignoring the context. This PSU project has a long history you're probably unaware of. Some of his questions have to be 'translated' to be understood in this context.

>> No.1526878
File: 1.48 MB, 3264x2448, IMG_20181031_160030.jpg [View same] [iqdb] [saucenao] [google] [report]

hey guys, can anyone help me with identifying a chip from a pc power supply? it is a cooler master 500w model with blown NTC thermistor and fuse (in pic related i already removed them). but the chip is what interests me. i tried googling but have found nothing
i want to know how can i test it in order to confirm its good and just replace those components
it looks really generic, like every psu has the same chip so i guess the pinout is the same for all
so if someone can identify it/them in general, i would be very thankful

>> No.1526884
File: 2.60 MB, 1580x703, Untitled.png [View same] [iqdb] [saucenao] [google] [report]

Not the guy you replied too. If you are refering to the brmonkey making a psu, it`s me and it`s already done. All I need is to find a nice case and decide on what pass device I`ll use (I`ve burned the ones I had bought by accident).
>things left to do
>settle on R values for fine adjust
>adjust loop gain so it can go all the way up to 26V (it`s 22v now)
>PIC code for displays
>SOA tests
Pic is me testing it in a universal motor. (Quite good tests loads btw, most of them are like 1kW and start spinning at low currents)

Next time be polite and answer what the guy asked for, or don`t reply at all. I mean, just a current source, who cares if he is doesn`t know much english or did not posted his circuit.

>> No.1526889

>answer the question or you're retarded
yeah no, if you want answers unconditionally then open your wallet. i didn't answer that question last night because i was drunk but i'm not even convinced a current source is what he needs. nobody calls a current source a "linear psu".

>> No.1526900
File: 905 KB, 2809x1000, Overview.jpg [View same] [iqdb] [saucenao] [google] [report]

To be honest I don't often post here, but thought it would be a good time to do so see if anyone is interested in what I am doing.
I have a few projects going on, the first being a JBC soldering iron controller, which is not too uncommon, however all other examples on the internet use regular transformers with multiple taps. I chose to try and design one with a switch mode powersupply at the input.
Technically the lifetime of the tip would degrade due to DC electroplating effects, but I doubt these are more significant than the corrosive effects of (especially lead free) solder and flux.
For the user interface I chose to be lazy and use standard encoder and 7 segment modules. The display module proved to be more of a pain in the ass to get working, since the chinese chip on there has a super shitty datasheet to use to bit bang the interface.
The electronics and software I have got working now, and currently I am waiting on a 3D printer to arrive to print the case.
The Rev 1 PCB had gotten a bit messy due to a mistake in the timer pins of the STM, and forgetting to add reset to the SWD connector. Still, 4 bodges on the circuit board doesn’t seem too bad.
Next version I will forgo the switching regulator and go for a linear one for simplicity (current draw of the 3.3V line is lower than expected, because of the 24V input I chose a switching regulator to minimize wasted power).

>> No.1526902
File: 454 KB, 2149x700, PCB revB.jpg [View same] [iqdb] [saucenao] [google] [report]

PCBs were ordered from Elecrow, fairly happy with them. At least a lot better than in the good old days where you had to make prototype PCBs yourself because ordering one was just simply too expensive.
Nowadays it costs more to make them yourself, than to order them from China.

>> No.1526906

He did not, you are just stupid. Linear psu= big regulator, it needs a minimum load to work properly, what is usually done is to use a current reference source IC at the output because they work over a couple decades of current and over a wide compliance range. What the guy asked was for a solution that did not involve purchasing an IC. Being not drunk does not make you not stupid. Read a book faggot.

>> No.1526910
File: 23 KB, 558x265, I_LED=200mA.png [View same] [iqdb] [saucenao] [google] [report]

Any capable academicist here? It's not about the 20Ω, it's the absurd result I get from this funny equation for Ud at 200mA.

>> No.1526915

Ud is a potential, it`s unit is V, not A. Calculating here Id at 200mV is 0.3uA, which is reasonable I think. And that equation is like ohms law for microelectronics, you should know it by heart.

>> No.1526916 [DELETED] 
File: 20 KB, 416x339, Capture.png [View same] [iqdb] [saucenao] [google] [report]

That funny equation is the basic diode equation.
It gives the current through the diode as a function of its forward voltage. Or of course the reverse. As you know this is a fairly steep curve, a small difference in voltage gives a huge difference in current.It seems that the point where the diode conducts here, the threshold voltage, is around 30mV. If that is the result you are getting, then I think that is correct.

>> No.1526918 [DELETED] 
File: 7 KB, 518x252, Plot.png [View same] [iqdb] [saucenao] [google] [report]

After reading >>1526915
I realise that the OP probably misinterpreted the equation.
I = current = amps
U = voltage = volts
For completeness sake, attached is a plot with on the y axis the current in amps, x axis voltage in volts.

>> No.1526924

it's suspicious how angry you are over people not answering someone else's question the way you wanted. why is that?

>> No.1526970
File: 7 KB, 220x178, file.png [View same] [iqdb] [saucenao] [google] [report]

I got a 2 transistor oscillator to work fine first try. I used FETs and added LEDs but otherwise it was this topology

>> No.1526976

Never mind, they were NPNs.

>> No.1527039

I just don`t like the pedantic stupidity akin to stack exchange a couple year ago. /diy/ and /out/ are the best boards and should not be touched by it.
there are some oscilators that use bjts on their reverse regions, they are tricky, kinda useless and usually don`t work.

>> No.1527050
File: 2.65 MB, 320x240, 1542509993946.gif [View same] [iqdb] [saucenao] [google] [report]

I don't get it.

Lithium ion batteries and supercapacitors use identical materials and have virtually identical construction, and yet one stores energy electrochemically while the other stores energy electrostatically. Both use the same lithium based electrolytes and carbon electrodes separated by a permeable barrier, yet they operate completely differently. How?

>> No.1527105

You can make super capacitors without lithium.

your question doesn`t make sense, it`s like asking why a solar panel doesn`t work with wind when a eolic generator does.

>> No.1527145

>bjts on their reverse regions
Oh I made one of those, only worked with a 2N2222, not a BC547.

Supercaps use an electrolyte contacting activated carbon to obtain a very high surface area. Lithium is a very lightweight ion with high solubility (therefore high conductivity), hence its use in this case. In a lithium ion battery, the lithium ions are changed from one oxidation state to another, a method which can store far more energy than electrostatic storage methods. The lithium ions are used for their lightness and high conductivity here too. Because no metal is coming in and out of solution, an inert electrode must be used for at least one of the anode/cathodes, and the choices for this are typically carbon or platinum. The latter of which is not used for obvious reasons.

>> No.1527146

You can also make batteries without lithium. I am asking how they operste differently despite being constructed in a seemingly identical manner.

>> No.1527151

What prevents the supercap from changing oxidation states and acting like a battery if all the hardware is there? That's what I don't get.

>> No.1527154

The construction is completely different. Look at a cell diagram of a li-ion, then look at the construction of a supercap.

>> No.1527155

I did and they looked the same, which is what led to me posting this question.

>> No.1527173

LiC6 + CoO2 > C6 + LiCoO2

LiClO4, could be a range of other electrolytes
Works on the action of ions from the solution being pulled towards each electrode, but they cannot contact it and cause a DC current to flow due to the inner Helmholtz layer. The use of lithium isn't required at all, you could use any electrolyte, and I doubt any of them would use cobalt. Though there are also pseudocapacitors, which do have some level of electrochemical energy transfer within them.

>> No.1527205

because they are not built in the same manner. One is
>conductive plate
>conductive plate
This is made to be really good at storing electrical charge.
the other is
>conductive plate
>conductive plate made out of something else
This is made to be really good at redox reactions

This is where engineering diverges from physics, that is why it`s simple to make a pico farad variable cap for a radio or a battery with coins at home but when you have to make them in millions, to produce and sustain high voltages or be very cheap things get complicated
>battery engineering = black magic (tried being a intern in a electrical material lab at my uni, they are wizzards)
>capacitors not so much

In theory a capacitor does not need an electrolyte, in practice to get large capacitances in a reasonably cheap and small device one is used.

>> No.1527236

I bought some christmas lights that flicker really bad. Taking it apart, each light is an LED in series with a resistor, and there are 25 lights in series operating at 120VAC, 60 Hz. There is not smoothing or rectifying other than what the LEDs are doing.
Would adding a proper rectifier be a fire hazard in a christmas tree? The current is about 8mA, and it'd be about line voltage once rectified (~170 volts).

>> No.1527249


>> No.1527251

This is great.

>> No.1527252

Hey fags. I have an inductor and a capacitor in series and they are not 180 out of phase. What might be going on?

>> No.1527255 [DELETED] 

I assume 'they' are current and voltage waveforms? They should be 90 degrees out of phase. An inductor adds a phase shift of at most j, which is 90 degrees. If it could add 180 degrees it could go unstable.
Either parasitic elements or your test frequency is not high enough.

>> No.1527256

>they are not 180 out of phase.
What is 'they'? Voltage? Current? Where? post a schematic.

>> No.1527257

Good morning and an interesting 2019.

>you should know it by heart.
No. The supposition of basic knowledge (like Ohm's law etc.) has been silently removed from OP's 'Rules' at the end of last November, but it still says 'Read posts fully. Solve more problems than you create.' You did not understand the problem and made up your own instead.

The question is about finding a value R such that I*R+Ud-5V=0 where I is given as 200mA and Ud has to be calculated using the equation at the top of the image. I did this and selected case B (R=20Ω), but the value of Ud I found makes no sense for the voltage across a LED operating at a current of 200mA.

Maybe some knowledgeable anon can explain this.

>> No.1527259
File: 23 KB, 252x220, eugh.png [View same] [iqdb] [saucenao] [google] [report]

this diagram just gets worse and worse

>> No.1527260
File: 20 KB, 416x339, Capture.png [View same] [iqdb] [saucenao] [google] [report]

I agree with the reply that the way the question is worded, is very very vague. First of all the reply is right that the 'funny equation' is very important, it is the elementary diode equation.

How I would solve the problem is as follows:
The Ud is given by 5-Ur = 5-I*R
Substituting this gives the equation:
I = 10*10^-10 * (e^((I*R)/0.025) -1)
which can be solved for different values of R:
2 ohm -> I = 240mA
20 -> 21mA
200 -> 1.8mA
2k -> 0.15mA
From here I can see that your chosen R=20 definitely does not give the 200mA current you are looking for.

>> No.1527261

What's it from?

>> No.1527262

my frequency divider that i've resumed work on for the first time in 2 years

>> No.1527265

>revisiting old schematics
oh the joy

>> No.1527269

>linear regulator
they're not exactly easy to find in small form factors and 24V Vin ratings
but berry nice work
>feels hot man

>being this pedantic
stay mad

hierarchical sheets are your friend

>> No.1527274

>hierarchical sheets
I do have 10 of the same circuit repeated, does that help?

>> No.1527275
File: 9 KB, 405x317, th.jpg [View same] [iqdb] [saucenao] [google] [report]


Schematic and the voltages are not in phase I will load my scope image next. I am not using a normal inductor it is a lower quality one of these https://www.pasco.com/prodGroups/coils-and-cores/index.cfm

>> No.1527276

>in small form factors and 24V Vin ratings
yes this too I have found to be the case. Still, linear regs are a lot less temperamental and expensive than switchers. And the switcher I have now is semi-exotic anyways. And worst part is, it only works when I remove the boost diode strangely enough. Still haven't fully figured out why that's the case.

>> No.1527277

What are the values you are using? Test frequency?

>> No.1527278

In Altium you can create symbols from sheets, very useful for repeated circuits.

>> No.1527279
File: 192 KB, 1274x754, not 180.png [View same] [iqdb] [saucenao] [google] [report]

What I am seeing on the scope
The capacitor is 10microfarads
and the inductor is what I am trying to determine.

>> No.1527284

do the transfer function, figure out the equation for the argument

>> No.1527285

So one of the lines is the voltage input to the circuit, the other the voltage over the capacitor?

>> No.1527287

the blue is the voltage before the capacitor an d the red is before the inductor
come again?

>> No.1527288

You only see a 90 degree phase shift, because you only see the voltage after 1 component. That means you have a +90 degree phase shift after the inductor.
If you try and figure out the transfer function from your input to your measurement point at the cap like >>1527284 mentioned you will see this as well.

>> No.1527291

You know, a = x_c/(x_c + x_l), where x_c = -j/(wC) and x_l = jwL.

>> No.1527294

i think i get what you mean.
i remember that now. at least i think i do.

what can i do to be able to see the resonant frequency? My intention is to add ch1 and ch2 to see the flat line but i am clearly doing something wrong.

>> No.1527296

>what can i do to be able to see the resonant frequency?
The impedances will cancel making it 'flatline' but that would result in you shorting out your function gen, not a good idea. it would be better to put them in parallel. Then if they are at resonance, you would just have a peak voltage.

>> No.1527299

Any good beginner tuts for making an adjustable strobe light, like the ones in the wooden box. Bonus points if it has a readout for strobe rate.

>> No.1527301

>being this buttblasted
Post a larger pic or state the problem then, I couldn`t read the ant font of the problem. Usually exponencial equations and inequations have no closed form solution, usually you arrive at the solution iteratively, or by using a simpler model. Look at this guy then>>1527260

I thought you misunderstood voltage and current or made a typo, regardless of OP ohms law and shockley`s equation are the ABCs of electronics, so make sure to understand their workings and nuances. Also if you are going deeper mass reaction law and boltzman law are also important.

>> No.1527303

Thank you.
>I = 10*10^-10 * (e^((I*R)/0.025) -1)
I would like to know
- Where does the first factor of 10 come from?
- If Ud=5-I*R, shouldn't it read (e^((5-I*R)/0.025)-1)?
Thanks again.

>> No.1527309
File: 15 KB, 1461x595, 1535496079142.png [View same] [iqdb] [saucenao] [google] [report]

absolutely! you can put the single circuit on a sheet and instantiate it as many times as needed. Pic related

>remove the boost diode
the freewheel diode you mean? maybe it's a shit diode. these things happen
still. very nice job. diamonds-making

probably not. but you'll just double the flashing frequency and maybe the duty cycle. add a smoothing cap and you'll probably be better off

>> No.1527316
File: 116 KB, 1362x224, y.png [View same] [iqdb] [saucenao] [google] [report]

Well then I think I've been wasting a lot of time.

>> No.1527318

bump, please help

>> No.1527326

that is always an unsettling possibility, isn't it
I haven't tried to split a device's units across pages though, maybe you'd have to import those back into the master sheet

it could be that nobody is having any better luck than you are identifying, let alone sourcing it
meanwhile, you could check the pass transistors to see if they're in shape. but tbqh most likely you're fucked

>> No.1527330

hopefully someone will help me

>> No.1527336

>Where does the first factor of 10 come from?
Read up on the diode equation
>If Ud=5-I*R, shouldn't it read (e^((5-I*R)/0.025)-1)?
Yes typo on my end

>> No.1527338
File: 25 KB, 341x199, Capture.png [View same] [iqdb] [saucenao] [google] [report]

>the freewheel diode you mean?
No, the boost diode. It lifts up the voltage to a point, such that when the highside fet is turned off, it can be used to drive the gate above the input voltage + fet threshold voltage (because the highside FET is N channel).
So technically it shouldnt even work, but it does.

And thanks for the compliment.

>> No.1527340

Searching "2D201" gives a power-supply-looking converter IC from Ricoh, might be what you're looking for but I doubt it.

I don't mind tearing the whole circuit up and starting again, since I've already done so like 3 times to replace components with pin headers since half my components will be mounted off-board. I could have them all board mounted and sticking through to the faceplate, but that's kinda a huge PCB I'd be using to get that, and I already have a bunch of cable to wire things upwards.

>> No.1527343

Wait, I come back to this since it is just plain wrong. I went through the calculations myself again.
It depends on the frequency what phase shift you will see. I wont spoil it for you, but go through the calculations and you will find a relation of phaseshift dependent on frequency. However, you should take into account that your impedance will drop very close to zero near the resonant frequency. Frequency gens dont like that.

>> No.1527350

Yeah I'd put a resistor in series with the inductor so you don't forget about the inductor's ESR and so you can use a known value that's much larger than the ESR to calculate the phase shift more accurately. Assuming that there's no ESR would probably give an erroneous reading, but knowing the resistance and doing ~5 tests at different frequencies should provide a good value.

Of course, the easiest thing to do would be to put any old resistor in series, tweak the frequency until you get a maximum (and no phase shift?), and then use that through f = 1/(w√(LC)) to find L.

>> No.1527351

>Read up on the diode equation
I did and found 1*10^-10.
Let's leave it at that.

>> No.1527357

oh, that one
weird, what reg is that? I've been looking through medium-Vin bucks and most seem to develop the bootstrap voltage internally, charge-pump style, via a diode from Vin to BOOST (usually internal). as it is, that 1N4148 would seem to provide a shortcut (red) for some inductor current, which you would prefer go through the load (green), especially if the load current is that small. with such a wide Vin-Vout, the Vin supply shouldn't have any problem at all making good Vgs just from Vin
I see what you mean about a linear reg like LP2951 making more sense here though

>> No.1527358

The particular one I am using is a MCP16301.
>that 1N4148 would seem to provide a shortcut (red) for some inductor current, which you would prefer go through the load (green), especially if the load current is that small. with such a wide Vin-Vout, the Vin supply shouldn't have any problem at all making good Vgs just from Vin
This was my thought as well. I expect that my boost capacitor has too low of an impedance for the minimum load, which makes the regulator screw up. From what I remember it instead just liked to put out around 1.something volt instead.

>> No.1527401

>ant font
Exotic language, it probably wouldn't have helped you but you can still validate the relevant data posted on the right side if need be.
>no closed form solution
The diode equation given has quite a simple closed form solution for Ud. That's why I was asking for a knowledgeable person able to check the value.

>> No.1527403
File: 391 KB, 1622x796, kysCAD.png [View same] [iqdb] [saucenao] [google] [report]

That's neat, I already want to die!

>> No.1527412

Did you make the smps too?
If you did do you have any resources to read on?

>> No.1527421

have you considered 74LVC1G inverters to make your routing life easier?

>> No.1527445

That's not a bad idea, but I've already bought my ICs and their sockets ready to install. If I do my job right in the layout screen, I won't need any jumpers and will have to do less soldering. The ~hundred resistors will prove to be far more annoying I thing.

>> No.1527447

too bad you didn't buy networks
oh well, multi-select and move will be your friend

>> No.1527458

The fuck man? Have you read their dsp docs?
The only good TI datasheets are inherited from NI.

>> No.1527459

>Have you read their dsp docs
No, but their transistor and IC docs seem to have the most extensive documentation that I've seen, especially when it comes to graphs.

>> No.1527463

Those sure are the same docs that national had for bog standard transistors and ics. Right now, Im looking at files DS008357 and SNOSBW5C. This is not the only case afaik.

>> No.1527478

Post values of all Rs pls

>> No.1527479

Use flux dummy

>> No.1527482

half of those pads don't look like they've been touched with solder. get some flux and go over it again

>> No.1527483

From which household item can I salvage a speaker which works with a 9v battery?

>> No.1527489

Bedside clock radio
An old PC speaker
pair of headphones
... or get one from ebay/china for $2
... or call a used/refurb computer store in your area and ask if they'll sell any old speakers for a few bucks

>> No.1527494

holy fuck bois I've went through numerous hobbies and occaisionally frequented general threads among 4chan but I'm actually doing ECE so I'm so hyped.. these threads random or pretty regular

>> No.1527496

I disagree, but maybe bc I don't gravitate to their parts. I love linear tech (hate analog) datasheets; for one, you won't find one without a sample schem (and not to mention those app. ckts with the std two resistor opamp lmao) and if we're talking aesthetics, there's no debate

>> No.1527509

Glad to have you aboard! It's a general, so when one thread dies another one pops up in its place. Now what part of this shitstained website did you weasel your way in from?

>> No.1527514
File: 103 KB, 1300x957, welcome to ohm now get out.jpg [View same] [iqdb] [saucenao] [google] [report]

they've been going for a couple of years pretty steadily now. it seemed to last longer than the /hrg/ ham radio general, fwiw

>> No.1527522

Well I started lurking /b/ since I was like 14-> uni jr. now. No boards in particular recently but im mainly a /fit/ & /mu/ vet. well desu I've never left gif but that's for yoU KNOW WHAT IM SAYIN

>> No.1527523

yeah makes sense with all the noobs + specificity
>posting gore in sfw
>that tip

>> No.1527534


This could be the gayest underage post I've ever seen in /ohm/. Is there a reddit thread causing this?

>> No.1527535

hopefully he'll get more autistic if he lurks /ohm/

>> No.1527540

I think they are either std or dime a dozen in shenzhen, either way there was one in a (possibly CM) psu I salvaged for some parts n heatsinks but if i remember right (although I was fucked up) there wasn't many ICs and I definitely think it's probably some dc-dc control. Unfortunately I don't have it anymore but if you could replace those components and power it w/out load (= output looks ok) then definitely probe, w/ caution, around those resistors for ~1.2 ref and the vouts, otherwise idk good luk

>> No.1527544

hi friend I don't appreciate the comment but frankly gay as an adjective would suggest the same for you. also I'm not underage and no reddit thread, all truth and in fact I was headed to gif when I noticed this thread on the homepage. its been so long I think >>1527535
is right

>> No.1527546

also wtf since when did they filter to be honest

>> No.1527548
File: 129 KB, 912x876, 1522168138948.jpg [View same] [iqdb] [saucenao] [google] [report]

>I'm not underage
>since when did they filter to be honest

>> No.1527549


>> No.1527550

feel free to translate that as "several years"

>> No.1527554

The only company where I have used a chip from, and which magically kept blowing up. They wouldnt answer questions because 'they dont do support for students', and 2 years after the fact send an email to everyone who purchased it through farnell that the equation for the compensation network was wrong.
I haven't, although i have made a SMPS in the past. Technically an SMPS is just a flyback converter. I remember I used a small 50mA 'regular' transformer to step down the voltage for the control circuitry instead of trying some fancy windings, and wound the main transformer myself.

>> No.1527562

it was a TI-designed chip and not one of their acquisitions?
tbf pajeet students of some castes believe that 'support' means someone else does their entire project for them

>> No.1527569

thanks anon

>> No.1527571

>it was a TI-designed chip
Yes it was, not an acquisition
>pajeet students of some castes believe that 'support' means someone else does their entire project for them
I know, they do the same everywhere on the internet. However this was for a student team as also explained in the email.
Since that experience with TI I avoid their products if I can find a competitor. Especially since there are quite a few manufacturers that did show amazing support for students as well.

>> No.1527592

>pajeet students of some castes believe that 'support' means someone else does their entire project for them

>post my project on instagram
>random pajeet starts asking me on direct messages why his own isn't working
>asks for links and youtube channels

Why are they like this?

>> No.1527617
File: 58 KB, 1369x775, octal transceiver.png [View same] [iqdb] [saucenao] [google] [report]

Can someone confirm that this is how the 74xx245 transceiver is supposed to be used?

I want to control some industrial stepper drivers with a 3.3v MCU. The stepper driver has optocouplers on every input which take between 8 and 15ma, so it would be like driving a bunch of leds and I don't want to hog the MCU's for current. Will using the octal buffer/transceiver IC help me achieve what I want?

>> No.1527619

What is even worse is the ones:
"Can you post schematics, code and detailed instructions how to build. thanks."

>> No.1527645

The wirng of the gate seems fine, but I've never seen a transceiver being used for what you want to achieve. Open collector buffers/drivers seems like a better option.

>> No.1527649

Usually for this stuff you'd use a buffer IC or discrete components as a buffer (since you dont need the bidirectionality).
however, it does seem like it would work, I am curious if this turns out okay, please keep us updated.

>> No.1527669

It looks like you aren't heating the pad. You need the iron to touch the pad and the lead until they're both above solder's melting point.

>> No.1527734
File: 68 KB, 1280x720, maxresdefault (7).jpg [View same] [iqdb] [saucenao] [google] [report]

The same components are in this board, so it works, but god knows if it's right

>> No.1527873

hey if it works, it works. Always fun seeing components being 'abused'. Or I prefer the term "Repurposed".

>> No.1527998

Brahmins don't believe in work, only the taking credit for

sure, that's one way to do it
or you could dial input current down to 5mA and still be within IC rating. the important thing is that you respect the current transfer ratio, which in the 817's case, is sub-unity, and adjust the receiver circuits on the output side if and as needed
>don't want to hog the MCU's for current
that's why it is customary for I/Os to pull loads down instead of up. thus almost no current from the I/O supply is drawn

are you sure it's an LS, and not an HCT or specific bus translator 74xx series?

>> No.1528016
File: 192 KB, 1310x1000, not awful.png [View same] [iqdb] [saucenao] [google] [report]

You know, with the rearranging of a few pins, this isn't turning out bad at all. I just hope I can fit it onto my 15x15cm square of double-sided copperclad. At the rate I'm going it will be 25cm wide, so I'll have to wrap the circuit in over itself or somehow cut the board in two. I suppose I should get a larger piece of copperclad on order huh.

>> No.1528022

actually the idea that I'll have to solder everything twice thanks to a lack of plated thru-holes means I might just get this done overseas instead. Though I guess the conductive-ink plating method isn't out of the question and isn't exactly expensive.

>> No.1528121
File: 100 KB, 921x640, 1545283995789.jpg [View same] [iqdb] [saucenao] [google] [report]

Where do I buy reagents for electroplating? Looking for a really powerful leveling agent, specifically.

>> No.1528155

How do I get boost converters to work properly? My circuit seems to work but I cant figure out the right values. I have a 3.3v output from a 555 but only get ~3.8v with the converter. Im using a 100uH coil and 20uF capacitor with a motor load that works at 3-30v. I tried 800uH and of course the voltage was higher but thats way too high im seeing circuits that end up with 30v and dont use such high values for the inductor.

>> No.1528156

Plain old copper plating I'd do simply in a conductive solution made of spectator/non-interfering ions, like sodium sulphate, and then use a copper electrode separately from the PCB itself. Amino acids could work for levelling (source: a single hackaday comment), but you'd be better off asking that question in SQT or on /sci/.

Rule 2 please.

>> No.1528157

not bad. just turn 'em 180 at the edge of the board and work your way
around circularly
random unorganized hints:
mind the sockets, those can be a pain to double-sided solder if they're the typical flush-mount cheapies, but not too bad if they're the machined-pin type
you shouldnt have to solder everything twice, only solder the side of the board that you are electrically using or where you need mechanical support against copper peeling
consider a slightly narrower trace for pin 8 of U3-U6, which you might could squeeze between R8/R9, R17/R18, etc. and the respective pins of the connectors
rotate U1 180 so that power polygons work for you instead of against you
there are still many options for miniaturization, if you've decided to send out for something already.
>get this done overseas
frankly, the drilling would have inspired me to do that already

by designing them properly and using proper components

>> No.1528167

>if they're the typical flush-mount cheapies
Oh shit they are, in that case I'll just solder them into the board completely. As far as soldering twice, the board would look better with it, but I'll pick and choose where I do so.
>rotate U1
I'll figure out U1 after putting the crystal and second inverter IC in place, until then I'm sorting out the shift registers.
>the drilling would have inspired me to do that already
I haven't given up yet. Mixing some powdered pencil charcoal with (thinned) house paint and applying it to the inside of the holes, then putting the thing in an electroplating bath before etching might just work. Plus I can test that before any of my etching stuff arrives. I'm still pretty determined to give home PCB fabrication a shot. Though I just figured out what a levelling agent is so I've got a while to go yet.

>> No.1528212

Li-pol to 3.3 dc-dc rated for 350mA?

>> No.1528217


>> No.1528226

Could you please provide me with some reference material for a 3.3v dc-dc converter that can hold up to 350mA, with input being 3.0-4.2v. Asshole.

>> No.1528234
File: 312 KB, 644x512, cunt.png [View same] [iqdb] [saucenao] [google] [report]

We're not your personal datasheet googlers, and this isn't some sort of headphone recommendation thread. If you want us to provide useful information the least you could do is be polite about it.

Second google result for "lipo 3.3V converter":
looks pretty decent if you want a PCB to solder onto. I think that's a fairly reputable site.

On Ali searching "buck boost 3.3V converter" I get this:
after sifting through some results that are distinctly boost or buck only.

Then a parametric search on digikey:
>min input voltage = 3V - 0V
>max input voltage = 4.8V - 100V
>max current output = 300mA - 32A
>voltage output = 3.3V
Gives me a fuckton of obscure SMT components like pic related and some high-power THT stuff, so probably not actually that useful to you. Perhaps Octopart would cater better to this sort of thing.

>> No.1528235

I now feel ashamed, thank you for being polite with assholes like me.

>> No.1528236

Just as long as you understand. /diy/ is one of the most well-behaved boards out there and I'd like to keep it that way.

>> No.1528252

>schizoid talking to himself in a slow board

>> No.1528253
File: 1.85 MB, 3264x2448, peen.jpg [View same] [iqdb] [saucenao] [google] [report]

fuck you and no I don't know how to use inspect element
saging for off-topic

>> No.1528258
File: 121 KB, 1080x421, Screenshot_20190102-153840~2.png [View same] [iqdb] [saucenao] [google] [report]

So the circuit in book goes like
Oscillator -> transistor -> 1k resistor, capacitor, 8 ohm speaker.

It produces a high pitch sound.
Now if I replace the speaker with 36 ohm (according to meter) earphones, what should the resistor value be?

>> No.1528260

can't you kids experiment anymore?

>> No.1528263

Would love to but earphones are a bit expensive. Sorry for being naggy here for the last 2 weeks but learnt a lot.

>> No.1528264
File: 85 KB, 717x527, morse.png [View same] [iqdb] [saucenao] [google] [report]

I am getting into oscillators, and want to make this simple AM transmitter. The problem is, I dont have the coil and want to wind my own, but I dont know how many winds. Looking on the datasheet of the 'BC Antenna' it says it is 200uH, but thats about it ( https://www.jaycar.com.au/medias/sys_master/images/9166372667422/LF1020-dataSheetMain.pdf ).

Google wasnt much help in how to wind your own AM transmitter coil. Any ideas?

>> No.1528267

first you need the ferrite rod
then you need to know its inductance factor A(L) which is usually expressed as henries per turn squared
then you divide the required inductance by the A(L) and take the square root

>> No.1528298

>replace the speaker
Leave the resistor as it is and connect another one in series with the headphone. The value of the added resistor will determine the loudness.

>> No.1528315

Thank you

>> No.1528373
File: 123 KB, 1024x768, HY-TB6600-Module-1.jpg [View same] [iqdb] [saucenao] [google] [report]

>or you could dial input current down to 5mA and still be within IC rating
But I can't count on the stepper driver couplers triggering with just that.

>that's why it is customary for I/Os to pull loads down instead of up
Some of them have a common anode and some have a common cathode, so I need to be able to handle both cases

>are you sure it's an LS, and not an HCT or specific bus translator 74xx series?
It's not a LS but I don't remember which are the random letters because I sold the board a long time ago. I'll need to search the internet.

So what I need is a buffer IC, hmm

>> No.1528459

the 74AHCT244 with Vcc=5V would be good. it takes 5V TTL level inputs (which are easily satisfied by 3.3V CMOS levels) and outputs CMOS 5V, maximum 25mA per output, 75mA in total. can be substituted with the 74AHCT240 to invert the outputs without changing any code

>> No.1528493

I want to mount my project in an altoids tin.
Does anyone know the dimensions of one?
Does anyone know where I can get standoffs that'd fit in an altoids tin?

>> No.1528504
File: 36 KB, 800x325, fm-receiver.jpg [View same] [iqdb] [saucenao] [google] [report]

What do I need to read to understand this circuit? I want to be able to make a circuit that reads arbitrary radio waves and displays the raw data from digital radio signals on a cheap LCD.

>> No.1528507

Bit of a brainlet question.
Is the DC output of a bridge rectifier equals to the peak voltage or the RMS voltage of the input wave assuming no load?

>> No.1528515

Idiot that can solder badly and do high school crafts class level ohm stuff but not tech oriented to learn much more outside of my hobby:

How hard and safe would it be to modify an existing power supply powered by an ac to dc "laptop" brick from 150w to around 300w and to modify output cables? And what would be the best way to start?

I'd would give it to be checked by an experienced person, before plugging but I'd have to have it done first.

>> No.1528535
File: 179 KB, 2048x1364, Dv5Acvge.jpg [View same] [iqdb] [saucenao] [google] [report]

>> No.1528538

Depends on the power supply. What do you have?

>> No.1528541


it'll take years. better just buy an RTLSDR

>> No.1528544

a small single board power supply from a small computer powered by ac to dc that connects by a 5.2 jack. There's not much on it, mostly some coils and capacitors, but everything is big enough to be soldered. So I thought it might be doable. I looked at cheap chinese "pico-psus" and they don't seem to have much more on them.

>> No.1528546

I can't understand the software for SDR either.

>> No.1528549

Wait, so where the does the ac-dc laptop brick come into play? I thought that was what's powering your power supply.

What about ICs? The more complicated ICs there are, the harder it'll be.

>> No.1528556

From my understanding that's coming from bad yt videos most of small psus don't do ac to dc just split the power coming from the power brick?

And there are no ICs visible. It's a cheap thing but at least it worked in a computer without a fault and was approved by an EU company so I guess it's working safely.

>> No.1528563
File: 96 KB, 550x550, 1545006804816.jpg [View same] [iqdb] [saucenao] [google] [report]

So uhh lets say i've got a 60W 12V PSU and a machine that has an average consumption of 50W but that often demands more, like 100W, for minutes at a time and if it doesn't get what it wants it malfunctions, so it wouldn't work just with the 60W. Is there any way i could make this combination work? Something to do with capacitors or batteries or something so when the demand is low the PSU can charge up the capacitor and when the demand is high the capacitor can provide the extra.
And yes i know i can get a 500W computer PSU for 10€ but in this case i do not want to do that.

>> No.1528571

>What do I need to read...
You would need to read a functional description of exactly this receiver, which uses a rather exotic principle to convert fm directly to audio by means of a slightly detuned oscillator.

>> No.1528577

Ok, what would I need to read to be able to infer what you just said from that diagram?

>> No.1528588

For minutes? Not for less than the price of a more powerful PSU

>> No.1528592

I don't know.

>> No.1528593

well what about 5 seconds? i think i could make it work
will probably just go with a bigger PSU if i can't find a reliable, cheap, and simple solution

>> No.1528595

What about a car battery in parallel to your "lets say" PSU?

>> No.1528608
File: 2.03 MB, 3024x3024, AD0D6AF2-F348-4401-AFEF-230A615F8363.jpg [View same] [iqdb] [saucenao] [google] [report]

This is kinda what I tried to do with my light box using the 12V/3A adapter and the 12V lead acid battery made something in the output side of the adapter start glowing red.

>> No.1528612

Let's see. An extra 50W at 12V is 4A. For 5 seconds that's 20C. We'll say you don't want the voltage to decrease more than 5%. So C = 20C/(12V*.05). So you need 33F of capacitance rated for more than 12V. That'll be tough. I don't know much about batteries but maybe that would work.

Or you could buy a better PSU and use the current one for some DIY stuff.

>> No.1528614

yeah i guess i'll just try to reduce the peak power draw a bit in my design and buy a bigger PSU

>> No.1528649

How would I make a neon indicator lamp be ON when a switch is CLOSED (mains is going to the circuit), instead of being OFF when the switches closes (and acts as a short, not letting any current to the lamp)

>> No.1528688
File: 12 KB, 168x300, images.jpg [View same] [iqdb] [saucenao] [google] [report]

So I've connected a small ws2812b matrix to Jinx! through a nano. Jinx! is also connected to Ableton Live, to make dynamic light stuff with the music.

Now I want to make a bigger LED wall do that, but big ws2812b walls get really expensive so I thought about p10 panels. The problem is I'm not sure you can connect a p10 panel to Jinx! through arduino at all. I've seen a tutorial to do it with FPGAs but they're expensive and I'm poor and doing it just for fun.

Anyone has done something similar? Or are there maybe cheaper alternatives to make big LED walls with similar communication as ws2812b? I have a raspberry too at hand, but for me it is important to be able to connect it to Ableton (windows laptop)

Pic related is not mine but it's basically it.

>> No.1528702

What the fuck is that?

>> No.1528710

Shorted Data, sixpack.

>> No.1528714

I'm so sorry. I'd recommend buying one of those $5 transistor testers off ebay or ali, they can measure inductance, as well as capacitance, resistance, transistor hFE, etc.

Well if you understand the working principle of each component (resistor, transistor, inductor, capacitor, amplifier) then all you have to do is piece the thing together from your knowledge. Plus look up about that method to turn FM into AM with a detuned oscillator, then the simple method to turning AM into audio.

Peak if you have a smoothing capacitor, RMS if there is none, and somewhere in between if it has a capacitor but is under significant load.

I wouldn't, it would likely involve replacing all the transistors, transformers, and probably some of the sense and driving circuitry too. Everything in those things are built to a cost, so you're not likely to find many parts that are only running at half their power capacity.

What does it do?

>> No.1528766

Old post, but you probably were just draining the current from charging caps. An NPN or op-amp used to buffer the output would probably make it work..

>> No.1528859

provokes thought

>> No.1528871

then you'd better find an introductory text for SDR and start experimenting
the ARRL would be a decent place for the radio newb to start, and they sell books for skill levels ranging from "what is a radio" to "can do Laplace transforms in one's head"

as long as there is always a load, wire the neon and resistor across the switch
the tiny amount of leakage current through the load and neon should be plenty to make the neon light up when the switch is open

look into Lattice iCE series, they're pretty economical and not completely incompetent. like $3 for a TQFP100 chip with 1000 logic elements, plus 50 cents for a serial flash to store the design, plus ~50 cents for an oscillator (if required). their IDE is a bit quirky but not terrible, and there is a fully open-source toolchain too. if you want one on a board, there are a couple for $25-$40
some of the lower-end Altera FPGAs are cheap too and can be gotten on chink boards for $10. their toolchain is a bit more traditional, some love it, some hate it

those transistor testers don't measure all that well down in the nanohenries though

y tho

>> No.1528879

How do I choose the right mosfet for a circuit? Say my load is a 2A high power LED and 5V source. Would I look at mosfets with a drain current higher than 2A or is there more important values I need to look at when trying to decide the right one?

>> No.1528880

look at the price anon

>> No.1528883

Well, when it comes to components I try to buy midranged price but surely I cant stick any random mosfet and expect it to work right?
I'm new to circuit building so picking the right components is probably the most aggravating part about all of this

>> No.1528885

Go on Digikey, open all the parametric search options, fill in everything according to your use case. After that, buy the cheapest remaining option.

>> No.1528888
File: 408 KB, 2478x1169, 1535021453103.png [View same] [iqdb] [saucenao] [google] [report]

apply safety margins to all inequalities below, and assuming clean ON-OFF switching:
1. filter by basic parameters: Id(max) > Iload, Vds and BVdss capability appropriate for circuit, Vgs rating suits the drive available. if you have a particular need for low Rds due to load characteristics, package size limitations, etc., also Rds(on) and maybe take a guess at Pd(max)
2. ensure the Rds(on) is sufficiently low at the Vgs you will be applying
3. check the Vds vs. Id and Vgs vs. Id graphs to verify that, given the Vgs you will be applying, it will pass the desired Id with an acceptable Vds drop
4. check that thermal and power dissipation limits will not be exceeded under operating conditions, Pd(max) < Rds(on) * Iload
other parameters may be relevant for particular applications, eg turn-on time, gate charge, body diode parameters, high-temperature parameters, safe operating area
Pic related, my usual jellybean n-MOSFET for low-power applications, annotated. which just happens to be suitable for your application

>> No.1528889
File: 32 KB, 1234x733, relay.png [View same] [iqdb] [saucenao] [google] [report]

dumb question:
Would it be better to set up a relay coil in parallel or series with a load? I'm assuming parallel but not sure.

>> No.1528890

you are correct, parallel
why: if you connect in series, neither will see the voltage (therefore current) it expects (Kirchhoff's voltage law)

>> No.1528935

Summer is here along with 100% humidity and day and night long 30C heat in the shadow. So I`m planning to make a ozone generator to get the funny smells and shit away. I was planning on using some sort of voltage multiplier thingy and to my surprise big clive made a bunch of articles on it. In the ones he builds, he says that capacitors rated for mains or ~600v can be safely used, but is it true? Why? because of the low currents?

>> No.1528942

Have a look at the air purifiers on Clive's channel, they're simple CW multipliers.

>> No.1528946
File: 59 KB, 1920x889, 1530477816237.png [View same] [iqdb] [saucenao] [google] [report]

in a Cockcroft-Walton voltage multiplier, it's because each component sees only a few hundred volts across itself (2x the input peak iirc)

>> No.1528961

i need a circuit that can produce a big flash for a short amount of time (think camera flash) and be a small as possible.
i've found really small circuits for Xenon lights that use small batteries (i'll use 3.7v li-ions) so i know it's doable, but maybe i can use LEDs and have a bigger flash ? no clue what to look for online, because searching "led flash" , "led strobe" etc. leads to tutorial on how to flash a led periodically. i just want the biggest flash possible for a small size. (like a 3cm*5cm pcb)

>> No.1528962

I want to make a small isolation transformer for very very little current (5 max) so I was thinking of using a iron nail or a small ferrite core. It`ll be used in mains voltage. My question is, how do I make the no-load current small? I mean, it still is a a wire shorted betwen mains voltage.

>> No.1528963
File: 503 KB, 750x748, cat.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.1528964

LEDs have a much smaller heat producing part, and so it can produce light for a much shorter amount of time before it overheats. Overdriving LEDs temporarily by more than ~200% of what they could sustain permanently is probably unheard of. Xenon flash globes are still king as far as temporarily overdriving. But if you don't mind shelling out for a 50W+ LED (not that expensive at all) plus some slightly more careful circuitry, you'd probably have a winner on your hands.

5mA? For the no-load current to be small, the inductance needs to be big. I'm afraid that means a fuckton of windings. But you can make winding a transformer like that easier by doubling over your wire (length calculated beforehand on like hyperphysics or something), putting the nail/ferrite rod in a drill, and going to town. Then cut the folded end and you've got two seperate windings. Nail is probably too small unless you use really fine wire, which would produce too much heat at mains. Make sure to wrap the nail in electrical tape so you don't scrape insulation off while you wind.

>> No.1528967

Thanks, I think I`ll spin it in a big nail, then bend it gently into a loop to avoid using a needle like instrument to wind toroids.(the tiny leftover airgap won`t affect things much I think.) Would limiting current with a resistor affect the operation too much? (to reduce # of windings)

>> No.1528970

>biggest flash possible for a small size
I think you'll want to stick with xenon, then. be advised that photo flash capacitors are not small, and neither will a driver circuit for a multi-tens-of-watts LED
Mouser might still sell xenon flash tubes new, otherwise check the surplus houses

with that tiny current, are you sure you even need isolation?
I was just now looking into some off-line buck converters. a couple of inductors, a few caps, and half a dozen resistors, and you can have regulated (but non-isolated) power for a burger or two

>> No.1528973

you probably wanted to search "led photo flash"
www electroschematics com/12394/camera-white-led-flash-illumination/

>> No.1528977

because safety comes first :)))
>buck converters
>half a dozen resistors
>few caps
I`ll stick to the iron and coil I think

>> No.1528987

Limiting the primary current with a resistor will cause the secondary voltage and current to be limited also. In an extreme scenario where 95% of the voltage on the primary side is across the resistor, the output voltage will only be maybe 10VAC (assuming unity turns ratio), and trying to pull current through this will cause it to sag even more. Needless to say, dropping that much voltage with a resistor tends to cause it to heat up a fair bit, and using a dropper capacitor would probably make things worse.

>> No.1528989

well, you have a point there. godspeed and good luck to you
post videos pl0x

>> No.1529013

Well I have a couple hundred feet of 28awg wire.. This is going to be a long afternoon.
Thanks anon.

>> No.1529037
File: 77 KB, 1200x800, DV__fSgVoAAR2Um.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.1529038
File: 128 KB, 1200x800, DvfWl7QUYAA4TRC.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.1529041


>> No.1529042
File: 25 KB, 266x285, 1543795832616.jpg [View same] [iqdb] [saucenao] [google] [report]

Anyone here design PCB for living???
Is 4k 32inch monitor benefit this task???

>> No.1529049

What does the batt option on my multimeter measure?

>> No.1529053


>> No.1529055
File: 500 KB, 1935x1290, 76BF5452-F025-4437-B625-424FA0D5962E.jpg [View same] [iqdb] [saucenao] [google] [report]

I was trying to figure this out with the $7 Walmart clearance one I got. The battery option measures voltage but puts some load on the battery. I’m not sure which exact numbers you’re supposed to be looking for, but I’m guessing that even if a AA is reading 1.6V with a regular voltmeter, if it drops to 1.2V with that small load on it, the cell is dead.

>> No.1529075

Why are you answering if you have no idea what the fuck you're talking about you cum gargling namefag?

>> No.1529112

So i got one of them RGB led strips rolls.. but now i don't really know what to do with it.. like i could put it under my work table or something but thats kinda gay.. maybe i could use it as an indicator for a new mail or some shit

>> No.1529169

I am looking to buy an oscillscope. I would like
to keep it under $1,000, but I really don't know
what kind of bandwidth, channels, etc. I need

Does anyone have a recommendation?
Should I be looking for analog, digital?
Any tips or places to look around at?

>> No.1529171

>nealy done project
>put it in case
>does not work
>check all components and connectivity
>perfboard hell
>doesn`t work
>bad transistor
>change transistor
>doesn`t work
I might have to redo this piece of fucking shit on a new board, this thing is driving me mad

>> No.1529175

What's a continuity tester for?

>> No.1529179

>check all components and connectivity

>> No.1529184

Anyone have tips on shaping/cutting solid core wires for use with breadboards/pin headers? I like to do a lot of prototyping and breadboard builds and I always get upset because I'm super autistic about my wiring but can't find a good way to make sure they're cut just right and I don't crush/fray the insulation.

>> No.1529185

whoops, i glossed over that.
if the components are fine, and connections are good, you could have a problem with impedance, two counter phases eliminating each other.
If you are finicky, get a razor blade, cut insulation, pull back, cut wire with wire cutters.

>> No.1529186

How experienced are you with electronics? Do you actually know what you need a scope for? If you don't know what features you need I'm gonna assume not, but that's okay.

You want at least two channels for signal analysis, if you're just starting then you can get a Hantek USB scope with two channels/10MHz bandwidth for like $65 on Amazon. Hantek also has discreet scopes that are really good value around $250-$350 compared to other manufacturers, and despite being chinkshit their quality is perfectly fine unless you're doing some real nigga shit, in which case you'd be using a >$1000 scope in a formal lab environment anyways.

>> No.1529187

Ah, didn't think about using a razor over strippers. That's a good idea, combined with that and doing actual measurements I can probably clean things up a lot while only sacrificing a shitload of time. Totally worth it tho.

>> No.1529188
File: 18 KB, 450x450, 1543648175760.jpg [View same] [iqdb] [saucenao] [google] [report]

as someone using a pair of 1080p 24" to design for hobby work, I can't see how more pixels and more desktop could do anything but help

reactive wall backlighting for your monitor

Pic related will CHANGE YOUR LIFE

almost everyone loves the Rigol DS1054Z. it's very capable, and easily hackable to turn on the software features, which includes 100MHz bandwidth

>> No.1529191

I was wanting to scope to at least be able to check digital logic, but I would also like to be able to use it for old analog circuits like NTSC/PAL chroma/luma signals in S-VIDEO.

>> No.1529192

I like to roll solid-core wire in the cutters before actually putting pressure down. It usually cuts through the insulation without crushing it, but a razor blade will be more precise.

>> No.1529212

are plain buffers essentially digital diodes?

>> No.1529217
File: 11 KB, 534x286, asdf.png [View same] [iqdb] [saucenao] [google] [report]

if it matters, I'm trying to set up a basic circuit in logisim and I need diodes to do what I want to do.

>> No.1529221
File: 68 KB, 1000x1000, image.jpg [View same] [iqdb] [saucenao] [google] [report]

I'm looking to buy these sort of fully sealed laptop charger thingies i need them to be 12V and somewhere around 90W, i live in Europe and they need to be up to whatever EU standards that apply to this stuff, i don't wanna get blamed when somebody's house gets burnt down by some Chinese PSU, where can i get them for cheap? at first i'll probably be buying around 10 pieces, maybe more later.

>> No.1529227


I don't get it. What are these things? They're really well made and quite nice to look at but I have no idea what they're supposed to do.

>> No.1529232

You don't need to buy a fancy monitor. I do design work on my shitty laptop screen and it's a pain, but my circuits are just as functional.

Seconding 1054Z

Can someone fill me in on Multisim? I've been an LTSpice guy for years but if multisim can do digital circuits I'm interested. Does it have some sort of functionality where I can choose my microcontroller from a list, put in the code I've written, and simulate with that? How's the ease of use relative to LTSpice?

>> No.1529352
File: 9 KB, 334x38, u wot.png [View same] [iqdb] [saucenao] [google] [report]

By translating the twitter page:
I'm as confused as you.

>> No.1529354

I could imagine them being ornamental flash drives. Can you fit enough electronics to interface a memory IC to a computer in the little bit of PCB you can see plus whatever's in the jack? My guess is, yes, but not for cheap enough for these to make sense as fancy flash drives.

>> No.1529358

レジン工作する化学徒。カメラはEOS5DMarkⅣ。1/12,13にHMJ2019(in東京ビッグサイト)でレジン作品と本を販売するよ!作品とレジン工作のHowto本の販売はCreemaでもhttps://www.creema.jp/c/nickelxp/item/onsale … 作品のオーダーや質問はDMからどうぞ。
"Chemist focused on resin-based manufacturing. Camera used is a "EOS5DMarkIV. Will be selling resin products and books at Hand-Made Japan2019/HMJ2019 (in the Tokyo Big Site convention center)! You can buy my Howto book on resin manufacturing and related products at (insert link here). Feel free to DM me regarding products and orders."

>> No.1529369

You're talking about peoples' houses burning down and shit, a taiwanese anime geology forum isn't the best resource

Shows what I know

That's cool, keep it up. Those are some really good-looking products and I'm sure you'll have tons of buyers. But please don't spam here advertising

>> No.1529371

I don't own the twitter, I'm just translating their bio.

>> No.1529378

Am I correct that it's close to $200 apiece for these flash drives? I was interested until I put the numbers into a currency converter

>> No.1529383
File: 56 KB, 200x214, bbb.png [View same] [iqdb] [saucenao] [google] [report]

it's closer to about $128.
Fun fact, to quickly convert 円 to $, just shift the comma over one to the right. It won't be perfect, but it'll be good enough for a back-of-the-napkin estimate

>> No.1529384

I hope the crystals act like cat's whisker LEDs, but I doubt it. In any case, I wonder what crystals they are? Those are 0Ω resistors and that's a mercury tilt-switch, right? I don't like the idea of putting a mercury switch in resin since the curing resin could conceivably change its size and break the glass.

It's more like 130 USD, but it's still a fairly steep price. I'd charge no more than 70USD for those, probably closer to 45.

>> No.1529386

I think you meant to say "shift the decimal point two spots to the left".

>> No.1529420

It looks to me like everything besides the USB jack and PCB provided by Sandisk does nothing. But if the guy actually made LEDs in the style of crystal detectors that's $200 worth of novelty on its own.

>> No.1529436
File: 1.80 MB, 2575x2277, payloadd.jpg [View same] [iqdb] [saucenao] [google] [report]

Check my sweet ass muon detector. It's going on a rocket in June and flying 70km up.

>> No.1529448

Well they wouldn't be very bright (just a small pinprick of light in a dark room) plus you'd somehow need to make the contact spring-loaded inside the resin. I can't actually find anything by googling "cats whisker LED" or "light emitting cats whisker", even though it's well known enough to be a demonstration in an electronics lecture. I found this video: https://vimeo.com/105901435 but I think he's literally just drawing an arc from the fucking thing.

But here's an interesting forum post: https://4hv.org/e107_plugins/forum/forum_viewtopic.php?92235.post about how you can get two diodes in antiparallel, which arguably could be applied for an AC LED.

>> No.1529451

What is the detector sensor? A geiger tube?

>> No.1529453

Does it run Temple OS?

>> No.1529455

Who is launching the rocket?
Also, that bend in the ethernet cable is giving me the willies.

>> No.1529457

wiring is almost comically bad

>> No.1529465

How do I calculate how much power my circuit will draw?

I'm planning a rather complicated analog circuit using two 10x10 LED matrices.
One's going to display a waveform, the other is going to be a VU meter with filters to separate the input.

It's going to use at least 6 LM3916, 200 LEDs, whatever the fuck I'm going to use for the filtering (don't know how much power active bandpass filters draw), various other ICs...

it's a lot, and I want to know how much it'll take to be battery driven before I start making it.

>> No.1529469

Basically everything is going to pale in comparison to your LEDs. Estimate how many are going to be on at once from how you want the displays to look. Everything else is negligible.

>> No.1529478

You first need to flip the Shockley equation back to front to get:
Ud = 25mV * ln(I/10^(-10)A + 1)
Now you shove the current into the equation and get 535mV.
The voltage is fairly low, compared to what you'd expect from a LED, but the Shockley equation provided is missing the ideality factor, which wouldn't be anywhere near 1 for a real LED.
Okay, so the remaining voltage for the resistor is 4.46V, which divided by 200mA is 22.3R.

>being this retarded

>> No.1529488

to give you an idea, it's essentially going to be

stuck together, except both will use 100 LEDs. Of course not all of them will be on so I guess I'll have to take an average.

I took a naive average of the first LED set and found on average 36 LEDs to be on at any given time in the video.
A naive average of the second video shows about 5.14, times 10, would be about 51 lights on on average.

>> No.1529494

Ok. 1mA is perfectly fine for indicator LEDs. Assume 2V Vf. 2mW per LED.

>> No.1529504

Alright, great.
I'll be honest I have no idea how shit works when it comes to designing power supplies for my circuit.
How do I determine the minimum voltage I have to supply to power the circuit?

>> No.1529512

It's probably 5V, but check the IC datasheets.

>> No.1529572

no, they're not
you need to use the correct kind of gate
>inb4 muh 200-piece bag of 7400s

>ore memory

I banged a chick from Temple once. it wa bretty gud. that she knew VHDL made it that much hotter

don't size the supply for the average, size it for the max plus a margin

if you're in a dark room and using high-efficiency devices, sure
something tells me these are meant to be seen tho, so 10mA is probably better

>> No.1529606

How should I approach learning about and understanding TTL circuits? Should I even invest time into it as someone primarily working with higher level digital stuff?

>> No.1529609

Maybe if you have shitty eyes. 1mA and standard LEDs is perfectly visible in a bright room.

>> No.1529625

Anyone know how to plot arbitrary points on an LTSpice plot?

>> No.1529661

>23 amps
in that tiny little thing?

>> No.1529689
File: 799 KB, 2560x1920, 4 bit adder.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.1529691

Design deets?

>> No.1529692

pot it in clear resin and shove it up your ass

>> No.1529694

how else am i supposed to make it
on a 2d plebeyan board or in god's full 3 dimensions to maximize space efficiency

>> No.1529738

Hello, I don`t know if it`s helpful, but I shake a lot sometimes because o anti depressants, and it really makes it hard to solder now and then. Most tweezers I find for electronics are either paper thin and bend after use, expensive, have no grip or all of the above, combined with the cheap chinese helping hands it really is a pain in the ass. But I found out that non-disposable surgical steel tweezers they sell in medical equipement stores are 11/10 for electronics, and they literally have ALL shapes and sizes possible because surgery need some weird shit.

>buy surgical tweezers instead of the ones they sell at electronic stores, they are cheaper and better

>> No.1529742

or just lay off the antidepressants

>> No.1529745
File: 879 KB, 2560x1920, 6 bits almost assembled.jpg [View same] [iqdb] [saucenao] [google] [report]

5th and 6th bit and now i'm outta trannies
hurry up based chinks

>> No.1529795
File: 6 KB, 732x566, usbasp_circuit.png [View same] [iqdb] [saucenao] [google] [report]

I got an USBasp today, but the zener diode felt off (D1 or D2 on schematic), will it work without it?

>> No.1529798
File: 40 KB, 257x400, Philips_SHB7150_main.jpg [View same] [iqdb] [saucenao] [google] [report]

These headphones have the shittiest feature ever, how would I go about modifying it so that the Bluetooth doesn't turn off when I plug an AUX into it?

>> No.1529799

No. Solder it back on.

>> No.1529801

no I bet it's there just for show

>> No.1529803

I'll try, it's SMD tho

>> No.1529813

Thanks, finally. That's exactly the result that got me shockley'd. Have a file that tells the spells in return.

>> No.1529815

Solder the pads first, then use a little bit of elmer's glue from a tooth pick, put that little bit under your SMD and place it on the board, let it dry. Once dry, then use your iron to solder it back on.
Unless you have a hot air rework station, that makes it easier.
Or if you have tweezers.

>> No.1529821
File: 15 KB, 444x378, F7904833-01.jpg [View same] [iqdb] [saucenao] [google] [report]

SMD is easy to solder with an iron unless it's 0402

BTW, would a mica resistor be able to heat a plate up to 230ºC for reflow soldering?

>> No.1529823

Find out what "detects" the jack in there and get rid of it

>> No.1529840

It's probably some trace from the AUX jack that needs to be cut.

>> No.1529861

It could be a MCU input stuck low or high

>> No.1529865
File: 75 KB, 640x640, chinee qaulity.jpg [View same] [iqdb] [saucenao] [google] [report]

>fell off

>> No.1529868

when the plug is inserted into the jack it passes a voltage, it raises / lowers that pin.
isn't that how all those detection methods work?

>> No.1529883

Yes, but that pin can be permanently left high or low due to an internal short. If there's no visible cause, that is 100% it.

>> No.1529890

Very possible.

>> No.1529918

What are you making? Some kind of register I assume, is it a shift register? Soldering all the components together like that seems pretty suspect, shouldn't you at least be using glue blobs to help mitigate the possibility of a short?

>> No.1529948

oh, he just saw gislain benoit's clocks and got all excited

>> No.1529998

Either, but with some form of layout that can be understood.

>> No.1530003

Problem is, the same signal probably allows the AUX input to be amplified in the first place, so by having the BT be permanently on, you might end up plugging the AUX in and have nothing happen whatsoever.

>> No.1530014

if you're trying to ape that clock maker, 2.5D is probably enough. do try to find an aesthetic, though

>> No.1530055

what's the go-to op amp IC?

>> No.1530060

lm393. it's trash but that doesn't matter in a ton of cases. tl074 and ne5532 are also common, where the former has high input impedance and the latter has low input offset, noise, and higher bandwidth. not sure what rail-to-rail op amps are popular.

>> No.1530061


>> No.1530065

Good luck dealing with all those oscillations :^)

>> No.1530069

for general purpose work, the LM321/358/324 depending on how many you need in a package. LM741 is basically obsolete. for specialist applications (high speed, low power, CMOS input, rail-to-rail input and/or output, beyond-the-rails input, shutdown pin, etc.), choose a vendor/distributor and work their parametric selection guide

that's a comparator
in a 5V single-supply system, TLV600x and MCP600x (same thing) are reasonable choices

>> No.1530072


comparator, op amp, what's the difference really

>> No.1530077

I prefer Burr-Brown over Analog Devices myself, but it seems the other anons are shilling them quite a bit.
You didn't say what application though.
One for audio is going to be way different than one for a line buffer or filter.

>> No.1530079

any pretense of linearity, for one thing

>> No.1530081

VCO/VCF/VCAs for synthesizers

>> No.1530083
File: 61 KB, 697x365, LM392.png [View same] [iqdb] [saucenao] [google] [report]

See for yourself.

>> No.1530094

you sure you don't want an OTA for VCx stuff?
there aren't that many types left but they are still in production. the LM13700 may be the easiest to get
or you could just say fuggit and grab one of the CEM chip reissues to do the heavy lifting, if you dig that sound

>> No.1530130

I'm trying to find some good resources for this does anyone have any recommendations: I have a project I can't solve through a raspberry pi or other boards like it. So I want to use this as a chance to learn making custom pcb and a microcontroller system. If I wanted a high power micro controller like an ARM processor, is there a guide or book anywhere that covers the basics of how to hook these more advanced processors with features like us, crystals, etc...?

>> No.1530140

ARM core processors aren't microcontrollers, they're discreet CPUs and generally require external resources to function.

What is your objective, how fast does it have to be? Connecting microcontrollers from MC or Atmel is stupid easy, the datasheets are like reading a LEGO manual and most times you won't have to build any discreet circuitry besides the power supply if it uses AC.

>> No.1530141

>raspberry pi is ARM
>can't do project with rpi
>needs to make custom pcb with ARM
IDGI anon.

>> No.1530142

I hardly know what the hell I'm doing, I wouldn't be able to tell the difference between an OTA and Op Amp anyway, I just know early synths used op amps for their amps/filters/oscillators so that's what I'm aiming for

>> No.1530151

you should look up some modular VCx schematics, and see what they're doing. also worth looking up for background is Nuts & Volts' series on "understanding & using" op amps and OTAs
it wasn't quite like digital logic where everything can be made out of NAND gates. analog designs usually include carefully chosen components and more than one weird trick. there are entire books covering the art of synth design

>> No.1530161

What I want to do is just read in data from four USB ports, and then based on that info play a sound file to an external speaker with up to four speakers playing sounds simultaneously. And I figured it might be better long term to learn and make a board with my specific need for 4 usbs and 4 3.5mm jacks to reduce on size and teach me how to do things.

I can't get the combination of peripherals from a raspberry pi, the processor has the processing power I need, just not the connections even with Shields and hats.

>> No.1530170

the STM32 series called and demands recognition as a microcontroller

>4 USB ports
host or device? what kind?

>> No.1530174

Device I think, it would be hooking up to 4 digital scales. Nothing more than 2.0 needed. And then when a scale goes over a certain value the sound is played. Which is why I need four separate audio playback options that can play different sounds simultaneously which is my main limiter from rpi. Can't get 3 extra audio ports onto it.

>> No.1530177

sure you can, you can add usb-audio dongles with however many channels you need all day long. frankly, for dealing with devices, the pi is probably the way to go. USB code can be tricky and yucky esp if you're writing the drivers from scratch (as you probably would be, talking to four devices)

>> No.1530179

That's a fair point, thanks for all the help.

>> No.1530188

what are FPGAs for?

>> No.1530190

What kind of sounds, like a klaxon?, and why wouldn't the GPIO from rpi be enough to flip some transistors to sound different frequency buzzers?
For peeing. https://www.youtube.com/watch?v=eB8sG4smWbo

>> No.1530243

bro you can add however many USB connections you want, the raspi has pinouts for its entire serial bus

>> No.1530283

Whew, just spent 2 hours figuring out that the problem with the stereo wasn't the amp or the speakers, but the cables between them. Re-terminated them, plugged the amp into the TV through an RCA-to-3.5mm cable, and now I've got better bass than you could ever get through those tinny TV speakers. Never use the stereo these days anyhow, the amp is a Japanese ~1991 Rotel RA-820BX4 and I'm quite surprised at how none of the caps are bad. The only dodgy parts were the 2W power emitter resistors that had their epoxy crackling off, but underneath that they were fine.

>> No.1530368

can i check if a peltier module is bad with a multimeter?

>> No.1530383

Yes. Never had one, but you should find that (making sure both sides are at the same temperature, possibly measuring in both polarities) the resistance is very low, and that you get a small voltage when heating up one side with your hand. The thing's datasheet should have quantitative values for both these tests.

>> No.1530390

How to choose base current limiter resistors for power aplications in bjt? The load is variable, and I can`t trust hfe, and I need them to give max current (there is other collector current limiting).

>> No.1530398

>can`t trust hfe
Can you trust your multimeter?

>> No.1530399
File: 307 KB, 1508x1136, getting there.png [View same] [iqdb] [saucenao] [google] [report]

Freq-splitter / retarded PCB making anon here, I'm almost done designing this thing and it will fit on the 150x150mm piece of double-sided copper I have here, thanks to the suggestion of wrapping it around on itself. I just followed what made sense for what modules were on each side of the inverters, which I think ended up looking all right. The ground fill is a fucking mess though, had to run a bunch of extra traces to carry ground continuity from one zone to another, though I arranged some jumper thru-holes such that I didn't need any vias. Now all that's left is to finish wiring up the clock circuitry and pretty it all up.

Assume a minimum hFE. 50 or so usually does it, though you may want to assume lower if your transistors are particularly wimpy. Then take your maximum possible collector current, divide it by this minimum hFE, and you've got your base current. Or, you know, use a MOSFET.

>> No.1530404

>measuring Hfe with multimeter
>it varies with current and temperature
top kek

>> No.1530409

Is there a soldering pastebin?
Looking for uses of different types/watts

>> No.1530426

>This is going to be a long afternoon.
How's the nail coil transformer?

>> No.1530436

Oh m8, hi. It isn`t going. I had a bunch of misfortune (bike brokedown, lost my housekeys, no fucking sleep at all, sick) I`ll probably start working on it next week. Found some retangular transformer steel cuts that could be bent into a toroid, probably that`ll do.

>> No.1530480

Is something like this good enough for a beginner : https://www.flipkart.com/aptechdeals-beginners-6-1-economy-soldering-iron-kit-electric-kit-7in1-25-w-simple/p/itmf6hdvrhyqepqy?pid=PHTF6H97YYXKASMG&start_url=BrowserLaunch_AMP&srno=b_1
Also do I really need to get a 15w and 30w iron as my book recommends or just get the 25w one above?
Many thanks.

>> No.1530494

it's no good but if the ~$4 pricetag is a limiting factor then that's probably the only option you have. if you can't afford a temperature controlled station (~$35 chinese) then there's no sense buying two soldering irons as a substitute. one 25w will be fine.

>> No.1530508
File: 20 KB, 528x482, 1527780293414.png [View same] [iqdb] [saucenao] [google] [report]

y not tho

>carry ground continuity
continuity is not enough, you need solidity. the more and fatter your ground connections, the better protected you are against glitches etc
compounding the problem, I don't see any bypass caps on your logic. gonna use air wiring for that?
still gotta watch out for those acid traps
yes, acid traps are gay

>> No.1530513
File: 271 KB, 1200x1200, 1535516833448.jpg [View same] [iqdb] [saucenao] [google] [report]

no but there's a bin of soldering paste just to your left
for board-level work, 60W is almost always enough. if you're applying copper braid to ground planes, a bit more juice might not hurt
if you don't have a 60W station with good temperature control, you have bigger things to tune first
Hakko's page on tip shapes is plain and informative
everyone seems to love the Hakko FX-888
in chinkshit, I'm a fan of Aoyue

>> No.1530520

Have a good time drilling all those holes :^)

>> No.1530531

Because I only have power fets made for switching applications, I don`t know how they`ll fare in a 60V regulation thing.
>ordered 50 tip121 instead of tip127 for a project because of dyslexia
kill me

>> No.1530537

they'll probably do pretty okay, actually, as long as they're rated for the voltage and of the proper pola-
yep, you're fugged

>> No.1530560


Thank you
Decided to cheap out and buy an expensive one later as I'm low on rupees

>> No.1530583
File: 834 KB, 2560x1920, 8 bit adder.jpg [View same] [iqdb] [saucenao] [google] [report]

based chink delivered the trannies a week early
and it works, fuck you haters

>> No.1530607

you should see if it will run off 110 or 220vac

>> No.1530611
File: 16 KB, 702x347, usb data lines.png [View same] [iqdb] [saucenao] [google] [report]


the other anons are wrong. the zeners are only needed if you have some malfunctioning device thats putting more than 3.6V into the data lines, which is outside of the spec. this protection is not needed in almost every case.

>> No.1530635

but the AVR itself could be putting more than 3.6V into the data lines in normal operation. thus, the resistors and zeners

>> No.1530682

how is that clusterfuck not shorting out

>> No.1530683

You ever hear of static electricity?

>> No.1530704
File: 14 KB, 495x296, 1542759768288.png [View same] [iqdb] [saucenao] [google] [report]

you ever hear of protection diodes?

>> No.1530705
File: 668 KB, 2560x1920, good thing i made it an arch so it can support more weight.jpg [View same] [iqdb] [saucenao] [google] [report]

i'm surprised how rigid it is
even held horizontally(pic) it doesnt sag unless i shake it around quite a lot

and now come the registers

>> No.1530707

>even held horizontally(pic) it doesnt sag unless i shake it around quite a lot
ah to be 14 again

>> No.1530714

So I'm going to try my first microcontroller build and I want to make a small package for loading just no more than 1gb of pictures and then showing them on a connected screen. I could use other things like an rpi but that's more than I need and I want to learn how to make the smaller package. So my question is does the following list cover everything I would need to map in like eagle ignoring specifics: microcontroller, usb port, HDMI port, power, external crystal, and flash nand memory?

>> No.1530734
File: 474 KB, 450x253, 1541508682788.gif [View same] [iqdb] [saucenao] [google] [report]

>flash nand
probably optional, if your program is small enough. you might be able to fit all the code onto serial flash if you have RAM
definitely add some DDR RAM
>first microcontroller build
more like a tenth microcontroller build

>> No.1530756

Yeah, that's what those diodes are for, numbnuts.

>> No.1530762

obviously you haven't heard of them, because the ones in my pic are already inside the micro and will do quite nicely to suppress ESD especially with those resistors helping out. please join the 21st century

>> No.1530781

>please join the 21st century
I bet you smugly smirked when writing that.

>> No.1530821

>bypass caps on your logic
Are those a necessity? If so, adding them to my circuit now shouldn't pose too much of an issue. I saw there was an EEVblog video on bypass caps recently, is that a good one to watch? I'll also have to look up "acid traps" because this is the first time hearing about them.

I have a drill press, it won't be awful.

>> No.1530827

Are polymer capacitors bad if they just smell funny? I might have heated one up too much when I was heatshrinking a wire and now it smells. It didn't bulge or anything and the farads and resistance still seem fine. Can I still use it?

>> No.1530841
File: 637 KB, 2560x1920, 1 bit memory.jpg [View same] [iqdb] [saucenao] [google] [report]

whats wrong with storing bits like this
it keeps the value when set is low and sets it to input when high
do note input is off when it has a straight path to gnd because i dont know what the fuck im doing
could i connect 8 of these and have an 8 bit register?

>> No.1530845

you can answer these questions yourself with a simulator, you know

>> No.1530847

where is ``clear''

>> No.1530858

is there a simulator that runs well on an s3 mini?
what do you mean? setting to 0 should be input negative and set high

im trying to make a simple cpu, would making the registers more complex save me a lot of effort elsewhere?

>> No.1530865

>a 6 year old smartphone
some other user on here has a smartphone and seems to be able to sim basic circuits with it. have you considered trying any?
>would making the registers more complex save me a lot of effort elsewhere
this is the point at which you should design the instruction set or something so that you know what you'll need out of the logic cells and register file. have you ever implemented a CPU in less esoteric technologies before? if not, for the love of god do that

>> No.1530869

>s3 mini
Are you saying that an s3 mini is your most powerful computing device in 2019?

>> No.1530916

Is there another thread?

>> No.1530935

no, wait until page 10, jesus

>> No.1530957

I don't speak Spanish.

>> No.1530968

>input negative and set high
why didn't you simply try it

>> No.1530974

because he has 1000 transistors and zero clue

>> No.1530986

Is there such a thing as a nonresettable PPTC fuse? How do I make one? I just need some kind of fuse made out of plastic.

>> No.1530996

y tho

>> No.1531000

I'm doing some crazy stuff with 3D printed circuits for a research project.

>> No.1531003

no, there is no such thing
research what a fuse is and talk to some materials scientists

>> No.1531023

The point of a PTC is that it is resettable, I don't think non-resettable PTCs even exist.

>> No.1531024

If I can't get the carbon particles to stick in place in the open circuit position, is there perhaps some sort of conductive additive I could introduce that decomposes under high temperature? Just spitballing for ideas here.

>> No.1531028

Why are you trying to do this instead of using a normal fuse?

>> No.1531031

I can't print a normal fuse, and the application for how this is used is kind of exotic.

>> No.1531032


>> No.1531033


>> No.1531034

Oh, interesting. I could see resistors but how are you doing caps, inductors, and ICs?

>> No.1531039

You just embed coils and parallel plates. ICs are kind of outside the scope of this though.

>> No.1531079

>You just embed coils and parallel plates
embed a piece of fuse wire...

>> No.1531097

circuit idea
joule thief dead battery battery charger

you take old batteries that aren't 100% dead yet but not good enough for anything else and put it in the device
it then charges a usb power bank so you can eek out the last power from those batteries before throwing them out

>> No.1531117

I'm trying to do calculations to see how much decoupling capacitance I'll need on my PCB, but I can't find any datasheets resembling my ceramic disc capacitors that show ESL/inductance or resonant frequency on them. Any hints? Or can I just slap a 10nF and expect it to be fine? From what I can tell, I think the rise/fall time I'll be dealing with will be around 15ns, which corresponds to idk like 40MHz, so I just need to ensure that my impedance at that level is low enough and that I have enough capacitance to supply the necessary energy for each transition, which I don't think will be an issue.

Calculating the charge consumed in the changing current as one LED gets turned on (∆I = 3mA) and figuring out what the voltage sag is with different capacitors being drained by that much current over the 15ns (225pC), I get 45mV of drop from a 1nF cap, which I'm assuming will work just fine as far as ESL goes. But still, knowing what inductance I'll be dealing with would help. I'm going to assume that these timing calculations will apply not only to the 74HC14s, but to the 04s and 164s too.

>> No.1531137

is there any way for me to figure out what the cutoff frequencies of a passive RC bandpass filter will be

also, are passive circuits worth it or should I just go for active? I don't want to have to make a negative voltage rail on my circuit

>> No.1531152

How about using an LC circuit to get a better filter?

>> No.1531160

it's for an equalizer/VU meter so it needs to be in the audio range. As far as I know, LC filters operate well above that.

>> No.1531161

no they don't

>> No.1531172

Case in point, crossover circuits for speakers.

>> No.1531187

It needs to be solid state. I can't have anything melt.

>> No.1531189

So I'm looking to make a function gen(sq, tri, saw, sine) and with a relatively low THD sine, and range from ~1Hz - 1MHz. I have a 400kHz DAC, and I know I can get a DDS ic for pretty cheap, but mainly it just had me wondering how this was done in old pre-DDS generators. Did they use the smoothed triangle wave approach, variable L-C, or something else

>> No.1531203

In that case mix a PTC powder with your filament mix.

>> No.1531213

LC sounds like a good option, you'd have one trimmer cap and a series of coils that you switch in and out of use to change your range. For minimum harmonic distortion, an arrangement with op-amps making compounding phase shifts and positive feedback sounds like a good idea, and you could do so with just potentiometers and capacitors to make it easy to tune.

Also someone posted a 2-transistor sine oscillator that was way easier to get working than a Wien Bridge, which I'd quite like to see posted again because I'm a lazy manchild who couldn't find it through google.

>> No.1531228

hmm so I guess this is one thing varicaps are good for. Thanks, this a good starting point

>> No.1531256
File: 41 KB, 381x811, what in gods name is this shit.png [View same] [iqdb] [saucenao] [google] [report]

I acidentally my 5A bridge rectfier and I can`t wait to buy another one. I have like 50 N-channel mosfets. Would it be a bad idea to use them as diodes to rectfy mains? just wiring them as diodes, not using some fancy phase control circuitry or things like that.

>> No.1531262

they'll work fine but they have a higher forward voltage. also the mains can see funny transients so you really want 400-600v rated fets with a varistor in front of them.

>> No.1531302
File: 8 KB, 300x350, 1525457172184.png [View same] [iqdb] [saucenao] [google] [report]

for 5V logic, common practice is to add 100nF-1µF across each IC and 10µF at the power inlet and you'll be plenty fine, with margin
the only thing to sort of worry about is the inductance of your traces forming an LC resonator with your caps, so if you feel like being extra safe you should try to keep any LC resonators away from your clock frequency

you can simulate an inductor to ground (for low-freq filtering) with two resistors, an op-amp, and a cap, Pic related

large inductors are not necessarily cheap. for a VU meter it's almost certainly overkill

typically they used the Wien-bridge and variable LC with a light bulb to control gain
later on they used the smoothed triangle approach

Clapp oscillator?

>> No.1531384

>you can simulate an inductor to ground
is there a name for this kind of filter?

>> No.1531396

look for gyrator filter

>> No.1531406


I like this circuit as a constant current regulator, though it gets annoying if your VCC is less than the required gate voltage needed to push some amps.

>> No.1531410
File: 10 KB, 600x375, 1520946423739.png [View same] [iqdb] [saucenao] [google] [report]

caps on logic lines would bandwidth limit the signal. really no need, unless you have a specific analog reason to do so. For under/overshoot protection use a clipper circuit with 2 diodes.

>> No.1531425

High impedance ground return paths are a big no, increases radiation and since ac return current will travel the path of least inductance you might get some unwanted effects.
Drop the small ground traces and just add a ground fill to both sides of the PCB with stitching vias. That should be standard, plus having one side with a fill and one without the fab might add thieving anyways so you might as well make it useful.

>> No.1531440

Why aren't you using a conductive paint/ink? Ironically graphite powder mixed with something cheap like elmers glue or acrylic was the best conductor if I recall correctly. There's guides/experiments on which are useful. A crowbar/current limiter circuit would probably be better than a fuse, and you could even tune the current sense element by using a 3d-printed resistor.

Batteries have a habit of corrosively expanding when pushed beyond their normal chemistry.

An annoying thing with signal generators, you have to filter non-ideal components, wasting power, regenerate them, which requires tuning and circuit protection, or have your next thing be tolerant of the non-ideality. For small, audio-freq signals it's not bad, but once you're pushing RF and digital control it's really annoying.

>> No.1531447

he needs something that holds state across power-off
but physical changes e.g. melting aren't allowed
I'll chalk him up to being an annoying fantasist

>> No.1531456
File: 1.87 MB, 1844x687, plsnobully atleast I`ve tried.png [View same] [iqdb] [saucenao] [google] [report]

Rate my first project. I ran out of good lead based solder and found a big roll that just says ``Solder`` and the manufacturer`s name and it`s terrible.
It`s a linear power supply. It goes from 0 to 24 V, and I haven`t settled with a current capacity.
I saw some other anons here saying things about big TIPs 127 or some other chonky transistors but I only had BD140s here, I`ve used 4 of them, each with it`s own diode and base resistor.
The voltage reference is a 7812, which will be used to power 2 fans in the box. The yellow heatshrink is covering a inline resistor (I`ll install the fine tuning pot later) because I`m naugthy like that.

Next project is a Ozone maker thing.

>> No.1531459

Also, could not find any thin solid core wire. I think I`ll strip ethernet cables and use those because flexible cables are a piece of shit.

>> No.1531463

if it works then congrats. post schematic.

>> No.1531508

The "solder" may not have been rosin-core, which would explain a lot of shittiness.

A lot of salvaged wiring I've found to be aluminum or something else, which is truly horrible to solder.

A convenient source of nice solid-core wire is breadboard wire. Comes in different standard lengths, already straight and easy to bend in right angles, etc. Lots of colors too. I don't know the normal material, so if it's aluminum that'd be annoying, but it's probably iron/steel.

Looks like a bunch of salvage components yeah? Next time I'd suggest using more board, so you can mount it easier, and you don't need connectors so close to the edge. I do like how at least one side is mostly soldered straight. I personally need to use KiCad/whatever because my routing is atrocious. Perf board just has that habit.

>> No.1531518

Ok, I was mounting the heatsinks and by acident the wire from the filter capacitor (2200uF) touched one of the heatsinks and discharged shorted through it (the whole thing was disconnected from mains, it was just residual 30v residual charge) and damaged all the transistors. (I think the mounting pads are connected to the collector). The rest of the thing seems ok. Tomorrow I`ll buy new transistors..

>> No.1531641
File: 371 KB, 1500x1154, eh.png [View same] [iqdb] [saucenao] [google] [report]

I'm using 0.5mm traces everywhere and have caps on all the Vcc pins of my ICs. I also just rerouted a bunch of traces and added some vias for the +5V rail such that I could get a better ground plane. Then I'll probably add a decently sized electrolytic to the Vcc rail, which should give me enough results. I'll test part of the circuit on a breadboard (or two) before actually printing the PCB, which may or may not work at 4MHz.

>> No.1531643
File: 301 KB, 1260x994, better.png [View same] [iqdb] [saucenao] [google] [report]

oh that's not the most recent image version

>> No.1531655

if you want that to work the input transistor needs to be a jfet or something. There is a diode in the BJT, setting input low won't do anything.

>> No.1531671

when paralelling bjts with separate base resistors do I need ballast resistors at the emitter?

>> No.1531709
File: 1.29 MB, 3264x2448, nippon.jpg [View same] [iqdb] [saucenao] [google] [report]

My 2 450V 1000µF caps arrived today, the fuckers actually had a charge in them that made a spark when I accidentally shorted them. The spark looked like 20V or so. I plan on using them as filtration caps for a PSU. They say Nippon Chemi-Con on them, and their ratings were around 1.015mF, but I got them from ali so I've no idea if they're legit. Is it common to find proper Japanese brands on ali? They aren't vented, but the bottom insulator around the contacts feels like it might be natural rubber, similar to my Rubycon caps bought from a more reputable source. They're 35mm in diameter, 50mm tall, definitely my biggest caps, they feel like the sort of thing I'd use to power a railgun, though I've no idea what voltage those usually run at, I'm assuming it's lower.

I also got a PCB ruler, 5 no-name 400V 100µF caps, and some IEC sockets among other things.

>> No.1531714

I had a 12V wiper motor connected to a PC PSU. It worked fine last time continuously for 10-20 minutes.
Now when i turned it on, it worked for a minute or so then it stopped with a huge pop. The PSU is toast, cant check the motor if its okay or not.
I have one PC PSU left and i sort of not want it to die. What protections can i add in order to guard the PSU?

>> No.1531720

A fuse.

>> No.1531724

I'd check the current PSU for a fuse anyhow, since they tend to have something like that. It may be a thermal fuse bundled in with the flyback transformers.

>> No.1531725

Fuse is fine sadly.

>> No.1531726
File: 156 KB, 1428x522, lmao.jpg [View same] [iqdb] [saucenao] [google] [report]

You checked for a thermal fuse?

>> No.1531731

It has none, just a MOV close to the mains, it works.

I said my goodbyes to the last PSU. Im sort of more interested in the prevention of similar events with the new one.

>> No.1531755

You can't use a MOV without a fuse, it shorts the two rails in the event of a voltage spike. Without a fuse of some sort this would trip your breaker.

>> No.1531756

it depends on your application but in general it's always yes. If there's a difference in base-emitter voltages (there will be), that gap will increase as heat increases => runaway

>> No.1531891
File: 5 KB, 358x278, 1528521136168.png [View same] [iqdb] [saucenao] [google] [report]

couple threads ago I asked about an old Marantz which I can't power

I checked continuity for about an hour thinking it was something fucked around the power transformer circuit. Turns out it's just the power cable that's weak somewhere along. I feel a bit retarded but at least now I know I can fix it.

>> No.1531916

>setting input low won't do anything.
That's why I suggested to experiment. The b-c junction is not an isolated diode, transistors can work in reverse with a low B and limited voltage. When the FF is set (OUT=H), there's one Vbe at the emitter of the set/input transistor and if the collector is grounded the base current drives the transistor in reverse. The question is whether it is enough to reset the FF.

>> No.1531939

how much resistance do i need to have in a conductor if i want to use it as a heater with a car battery?
i tried to measure resistance in a heating pad but it only shows one ohm and that cant be right, it's too small

>> No.1531948

If you're paralleling transistors, presumably it means that one transistor can't handle all of the current. Yet without emitter resistors, there's nothing to force the transistors to divide the current evenly. And in practice, they won't divide it evenly. Whichever transistor has the highest gain (or lowest Vce[sat]) will get most of the current. This will result in it dissipating more heat and having a higher temperature, resulting in both increased gain and reduced Vce[sat], resulting in thermal runaway.

Adding an emitter resistor makes the collector-emitter current largely independent of the transistor characteristics, being determined by Vb=Vbe+Ve=Vbe+Ie*Re => Ie=(Vb-Vbe)/Re.

>> No.1531952

>it's too small
Why? Let it heat for a while, zero your ohm-meter and measure again while it's hot. 12V at 1Ω means 12V*12A=144W

>> No.1531953

V=I*R => I=V/R, P=V*I=V^2/R => R=V^2/P.

But note that the resistance of most conductors increases with temperature. Measuring the resistance when cold doesn't necessarily reflect the resistance at its operating temperature.

Realistically, you need to find out the intended supply voltage of the heating pad.

>> No.1531957

PM-710 up, sounds pretty good even though I'm no audiophile. I should still try to change the caps some day, just so that "full restoration" doesn't just mean fixing the power cable, dusting the inside and cleaning the aluminum front...

>> No.1532049

So i have esp32 which operates on 3v3 logic.

I can power it from usb or feed 5V into the VIN pin which goes through a voltage regulator.

But i wonder, will i be able to power it by feeding 3v3 into the VIN pin? will there be like some voltage drop on the regulator that will cause brownouts or will the esp32 automatically bypass the voltage regulator or some shit?

I am too afraid to feed the power directly to the 3v3 pin since that one is behind the voltage regulator so any slight voltage spike would mean bye bye my incredibly expensive $5 board

>> No.1532072

Might read the datasheet for that.

>> No.1532083

>will there be like some voltage drop
if you have a reliable, stable 3v3 source, go ahead but don't provide power from any other source (e.g. USB) while you do. if you don't have such a source, don't bother

>> No.1532085

I have freed myself from the confines of perfboard. Im now recognize the superiority of deadbug and point to point connections. The ancient ones were right. Thanks bob pease. Hail BJT adder anon

>> No.1532087

Would anyone here recommend buying a Tektronix 2465BDV for $800?

>> No.1532089

Too expensive.

>> No.1532094

By how much? Like $100?

>> No.1532096
File: 136 KB, 846x960, 1527589235077.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.1532105
File: 93 KB, 493x307, bob pease.jpg [View same] [iqdb] [saucenao] [google] [report]

Peasants will never understand

>> No.1532107

what does it matter if it is stable tho? the voltahe drop will still occur

>> No.1532159

By $800. Buy a scope that was made in this century. The Rigol DS1054Z is the default here

>> No.1532164

Don't like analog scopes? Ok...
Would spending another $150 on a siglent sd1104z-e for 100 MHz be all right, or is the rigol enough for most things like spi debugging?

>> No.1532177



>> No.1532188

Save the $150

>> No.1532242

why do you exist

Name (leave empty)
Comment (leave empty)
Password [?]Password used for file deletion.