[ 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: 2.29 MB, 3248x2436, based crt scope.jpg [View same] [iqdb] [saucenao] [google] [report]
1945940 No.1945940 [Reply] [Original] [archived.moe]

Thread let the smoke out:>>1937846

>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:
Don't ask, roll:

>Principles (by increasing skill level):
Mims III, Getting Started in 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)
Logisim (Evolution)

Mouser, Digi-Key, Arrow, Newark, LCSC (global)
RS Components (Europe)
eBay/AliExpress sellers, for component assortments/sample kits (caveat emptor)
Local independent electronics distributors

>Related YouTube channels:

>Li+/LiPo batteries
Read this first:http://www.elteconline.com/download/pdf/SAFT-RIC-LI-ION-Safety-Recommendations.pdf

>I have junk, what do?
Shitcan it

>> No.1945947

babby very excited about using a scope for the first time
kill yourself

>> No.1945961

No, I just took the pic from the previous guy.
And I've used my scope for over a year, and used scopes for years before that in labs.

>> No.1945967

What the fuck is wrong with you. Who acts like this?

>> No.1945998
File: 54 KB, 708x800, 1580872056423.png [View same] [iqdb] [saucenao] [google] [report]

>What the fuck is wrong with you. Who acts like this?

>> No.1946005

>There is nothing wrong with your television set, do not attempt to adjust the picture.

>> No.1946007 [DELETED] 
File: 117 KB, 855x1024, 1603689448635.jpg [View same] [iqdb] [saucenao] [google] [report]

eat shit nigger fag

>> No.1946012

If I wanted to get from total beginner to a more intermediate level, would it be a good idea to look up datasheets from ICs and build the sample circuit included in the datasheets?

Also, how do I learn basic PCB design to have my circuits sent to JLCPCB?

>> No.1946021

based, we /b/ now

>> No.1946026

>would it be a good idea to look up datasheets from ICs and build the sample circuit included in the datasheets?
i don't think so. this is equivalent to copy-pasting code off stackoverflow. it's an efficient way to work but you don't learn a lot by doing it.

>how do I learn basic PCB design
75% of it is learning the shitty interface for your software. the rest is ground planes, trace sizing, and not fucking up your footprints. best way to learn is by making cheap projects. if you can't think of anything then design a usb-uart converter.

>> No.1946048

>i don't think so. this is equivalent to copy-pasting code off stackoverflow. it's an efficient way to work but you don't learn a lot by doing it.

I respectfully disagree. One of the best ways to learn to code is to take some working code that does almost what you want, and then incrementally changing that code. Same way with a circuit. Start with something that works, verify that yours works like it should, then modify it and try not to break it while changing the overall behavior. Same principle applies to coding and to circuit design.

Or start from scratch, which nobody actually does.

>> No.1946051

Is there something like thingiverse but for electronics? Where ppl share their designs and files and other ppl can build them or have PCBs made?

>> No.1946056


>> No.1946057

Hackaday also have a project hub.

>> No.1946059

Uh nice! I just saw that PCBWay also has a shared PCB section with direct Gerber download.

>> No.1946060

that's boring, but it you have the autism go for it
think of a project you want to work towards, look up a schematic, break it down into modules and make each building block separately for practice

>> No.1946113

starting from scratch is imo the best way.

>"I made something completely and totally my own"
>"oh yeah i sorta tweaked someone else's design, that someone else has already proven and did all the risk taking for me"

like >>1946026 said, make simple cheap projects. JLCPCB is $10, components should not exceed $15 for simple projects. USB to UART converter is a really good one because it will be useful, and is straightforward.

Another good project is to take some obscure MCU, and build an Arduino-style devboard for it.

Be prepared to fuck up PCBs, redesign, and respin. Yeah its gonna cost you another $10 and another week - welcome to the world of engineering, mistakes have consequences. So learn how to use the Design Rule Checker, and don't rush it.

>> No.1946133

Have you lost your mind because the election didn't go the way you want. Is that what this is about?

>> No.1946178


>> No.1946203

its my fault. im the guy who posed the question about tesla coils. is Q more important than E. more stacks of lower voltage or higher voltage and fewer stacks. i posted the vid links and question about what made the difference and how could a guy with good maths (great scott) make such a shitty big tesla but the guy from weekend projects makes one that ruins on less than 50 volts and runs circles around greatscotts. i did it in 2 posts in the previous ohm

many sought to prove themselves and it decreased activity on the thread. then some found their whole life was a lie or ran into the dreaded diode breaks voltage as if you high side switched a npn / nchannel and no sparks. makes it hard to look at yourself when you fail at something thast basic math right

now they are disturbed. now they have gone through the metamorphosis . next time i rant about farads being lies and capacitor memory and even current making more of a difference than listed farads. then they smear shit on walls screaming witness me

>> No.1946235


>> No.1946243
File: 2.53 MB, 4032x2268, 20201106_205920.jpg [View same] [iqdb] [saucenao] [google] [report]

Hello, I am trying to fix my vacuum cleaner. The problem with it is that the roller is not activating like it is supposed to. There is a small switch that is depressed when the vacuum is in upright position. When you pull back the vacuum to start using it, the switch is released and the roller motor is supposed to start. Mine takes many tries to get the motor to come on, and then it will usually turn off immediately after. I have taken the vacuum apart and now I know that it is not a mechanical issue but an electronic one.

Do you guys think this issue could be fixed by replacing the switch? Or is it the motor that is malfunctioning? Is it worth it to try and replace parts on this or is it a lost cause? Thank you

>> No.1946244
File: 2.00 MB, 4032x2268, 20201106_205845.jpg [View same] [iqdb] [saucenao] [google] [report]

Here is the switch.

>> No.1946249

Measure the voltage across the switch when it is and isn't tilted, that should tell you if the switch is working or not.

>> No.1946255


first thing is see how the wires are attached to switch. if it's done with 1/4-inch spade lugs, then pull them out with pliers, and short out the two wires with a paper-clip, or something.
if the motor run fine each time, then the switch is bad. or bypass it and have the rollers run constantly.

if the wires are soldered, try another tactic.
motor's attached thru a connector (white plastic doo-dad) so, should be a simple matter to separate the motor from the circuit. (the connector has a little lever you depress before pulling out connector). once done, you can test first the motor, then the switch.

to test motor, take an old extension cord, or power cord from some dead machine, and cut the wires and strip them. insert power cord into the motor's connector to see if it runs. if it fails or stutters, you might be able to fix it by spraying contact cleaner into the brushes.

to test switch, connect light bulb to the place where the motor usually connects.

>> No.1946260

Thanks for the advice. I have now found out that the switch is going to be very difficult to remove, I think they glued it in there. So I'll have to deal with that first if I want to move forward. In the manual it says there is a failsafe stop for when the roller is clogged. So I'm going to clean it all really well and see if that helps. If that doesn't work I'll try out your ideas, thanks again.

>> No.1946310

Best way to learn is to pick a project and try to do it. If you have a bread board just start small:
>make sure you know how resistor works
>make sure you now how LED works, you are going to burn a couple of them by being a retard
>try making a transistor switch something on and off
>try to make a class A amplfiier
>try to amplify some signal with an opamp
>try to filter a signal with an opamp
>guitar pedals
>light stuff
>small power supply for yourself

>> No.1946311

I say this because unless you are reading some really good app notes and datasheets theyĺl probably be kinda garbage and not explain anything. Those circuits are more the building block you use when you have some experience instead of always doing circuit laws to figure stuff out.

>> No.1946312

>they make PCB-mount XT90 connectors
that will be 5000 ounce copper traces, thanks pcbway

>> No.1946315

Do measurements before doing anything that will be difficult to undo, such as ungluing a potentially fully functional switch. It's either the motor, the switch (or some other switch or interlock), or the main control board. You should be able to check where the issue is just by measuring the voltages on the circuit while it's plugged in. Some reverse engineering required.

>transistor amplifier topology before op-amps
I'd maybe put nonlinear common-base, common-collector, common-emitter stages before op-amps, but actually getting a class-A with reasonable behaviour is a bit of a nuanced process. At least it is compared to selecting resistor values and not having to worry at all about input and output impedances.

To be honest I'd only ever use a class A or B or AB if it's the output stage of an op-amp feedback loop. Op-amps are cheaper than the kind of power transistor you'd want as an output stage anyhow.

>> No.1946340

you are right. Op amps are easier to start with amplifiers than just transistors. To the guy asking the question, intuitive knowledge of what gain is and how it adds and subtracts is a very important thing. And you do not need to order a PCB done to fuck around, a breadboard will do fine.

>> No.1946341

I would actually encourage burning a couple LEDs, just to see what it's like

>> No.1946342

I have a guitar with mains hum. From what I can see, there are no gnd loops and everything is star grounded with cheap shielded wire. All grounds tested as shorted together.
When I touch the bridge, the hum quiets a bit. But when I touch the selector switch, it starts buzzing loudly.

Does this mean the pickup switch's common ground is the issue?

>> No.1946346
File: 3.62 MB, 4000x1846, IMG_20201107_122759.jpg [View same] [iqdb] [saucenao] [google] [report]

Greetings from /g/,
how can I go about finding a replacement of this remote?
We got it from the landlords quoting "if it doesn't work, throw it away and use the key".
I can't find the exact model, tried googling component numbers, but to no avail.

I'm afraid my wife will hit the wall sometime when trying to park next to the keyhole.

Can I get relevant information from the receiver itself somehow?


>> No.1946361

What's the deal with those jumper blocks at the bottom? They look bent.

>> No.1946362

they just look bent, they have something like extensions on top for easier plug/unplug

>> No.1946364

Maybe the button is worn out or the crystal got fucked up from dropping the remote.

>> No.1946365

both buttons seem to work - they produce a continuous beeping sound, the door does not open. I will try pressing them longer etc.

Now I question if this is the right remote for that opener.

>> No.1946380

If you have a frequency counter or SDR setup you can test the RF output. Other than that, you might need to reprogram the remote which is probably what the second button is for. See if you can get a model # from the opener itself so you can narrow down universal replacement remotes that will work with what you've got.

>> No.1946390

The model of your remote is Came MT2-C. Link below (translated) says that the crystal is drifting and needs to be replaced.


>> No.1946394

to add:
Telecomando TOP302M possibly

>> No.1946398

Nope. Came TOP302L.
Newer equivalent is JANE J-Q2 26.995 MHz.

>> No.1946402

JANE J-Q2 30.900mhz

>> No.1946406
File: 115 KB, 962x778, reference.jpg [View same] [iqdb] [saucenao] [google] [report]

Came TOP302L programming instructions.

>> No.1946423
File: 430 KB, 816x590, came.png [View same] [iqdb] [saucenao] [google] [report]

sadly I don't have any, nor I know how to operate.

>Came MT2-C
I also found this string on the wider chip to be somehow helpful, couldn't find the exact model number tho...? Will read this through tho, thanks!

both looks similar and have seen them on google, couldn't make my search any more exact tho

>JANE J-Q2 26.995 MHz.
>JANE J-Q2 30.900mhz
And I've also seen Janes as replacements for the old bricks like mine. Just not sure about the frequency

Thanks, my remote has 2x6 jumpers. I will have to read this couple of times to understand it.

Pic related is the receiver in the garage.
Btw, what is the difference between these big old brick remotes and the new small ones?

Sorry for the noob questions and thanks everyone!

>> No.1946426
File: 743 KB, 421x912, file.png [View same] [iqdb] [saucenao] [google] [report]

ah, just saw the silver component
>MEC 4k

>> No.1946427

>I will have to read this couple of times to understand it.
Take a pic of how the switches are set on the receiver/motor. Then set the programming jumper on your remote and push the buttons (left = off, right = on) to duplicate the switch settings on the receiver. Then remove the programming jumper from the remote and test. Note: if the crystal is drifting (bad) then it still won't work.
>what is the difference between these big old brick remotes and the new small ones?
Mainly operating frequency and RF comms protocols.

>> No.1946430

>what is the difference
I don't know if the newer ones can be programmed directly or if they only copy codes from the original which can be programmed.

>> No.1946434

I think I will try standing below the motor and pressing the key.
If it works, and I order another 26,995 mhz remote, it should be just about cloning, right?

>> No.1946437

>If it works, and I order another 26,995 mhz remote, it should be just about cloning, right?
Yes, but keep the original if you ever need to change the receiver code in the future. If it doesn't work you need to replace the crystal.

>> No.1946439
File: 90 KB, 247x133, file.png [View same] [iqdb] [saucenao] [google] [report]

Thanks, just was thinking about that it should be so hard to solder a new one.

Last question - which one is used to fine-tune the crystal ?
I guess the blue one right next to it.

What a great board, I should core more often!

>> No.1946441

Yes, the blue pot is for fine tuning. Adjust that if your remote only works if you're really close to the receiver.

>> No.1946475

>you do not need to order a PCB done to fuck around, a breadboard will do fine
100% this.
Unless you're spending megabux on shipping, the lead time makes iterative prototyping a long and tedious process that should only be reserved for projects that need a proper PCB. Even then you can probably get away with manhattan for higher speed designs so long as you've got the excess copperclad. Not sure about length matching though.

>> No.1946478

>Btw, what is the difference between these big old brick remotes and the new small ones?

new ones use tiny batteries plus surface-mount components.
beyond size, newer ones also have much better security: the old ones tend to use a single code, so anyone can listen to it, record it, then play it back to open your door and steal your garden hose. newer ones change the code after each use.

probably your best bet is to replace receiver and transmitter. this probably involves paying someone to do it, tho it's a very simple operation. you typically just have to match 3 wires: positive, negative, and output.

>> No.1946483

>>Btw, what is the difference between these big old brick remotes and the new small ones?
>new ones use tiny batteries plus surface-mount components.
>beyond size, newer ones also have much better security: the old ones tend to use a single code, so anyone can listen to it, record it, then play it back to open your door and steal your garden hose. newer ones change the code after each use.
>probably your best bet is to replace receiver and transmitter. this probably involves paying someone to do it, tho it's a very simple operation. you typically just have to match 3 wires: positive, negative, and output.

That's not an option, we are in a residential building sharing the same underground garage with the building next door.

>> No.1946486

>residential building sharing the same underground garage with the building next door.

then this is not your problem. just ask building management for a new remote, and pay some over-inflated price for it.

>> No.1946494

What's the underlying principle of analog drum machines? They don't seem to be oscillators. I'm trying to look up "analog drum circuits" but they only talk about the drum machines rather than explaining it.

>> No.1946499

I think they're oscillators that taper off quickly, at least for bass drum and such. Add some random noise (LFSR) to the oscillator for a snare, or just use the noise plus some envelope shaping for a symbal. Shorter envelope for a hi-hat.

>> No.1946517
File: 342 KB, 1881x1120, lainscope.jpg [View same] [iqdb] [saucenao] [google] [report]

a lot of the really old ones used resonant circuits, with an inductor and a capacitor. they would be fed a pulse to trigger the resonance and make a bit of oscillation.

>> No.1946533

I've never heard of LFSR before and I have some shift registers lying around so I'll try experimenting with those.

I've never made a resonance circuit mostly because I've never had a use for inductors. I've got some scrap appliance boards with some toroids I should salvage.

>> No.1946558

mostly not fed by pulses but by positive feedback loops, making the system able to send out consistent sine waves

>> No.1946583
File: 224 KB, 1280x1268, df0_df1_selector_switch.jpg [View same] [iqdb] [saucenao] [google] [report]

i was trying to install a selector switch mod into my amiga. in doing so i managed to damage the socket for the even CIA chip.
i'm pretty competent at soldering but i don't know what i need to search for to order a replacement.

>> No.1946587

will these work or no?


>> No.1946597

That’s a very standard socket, if you search “IC socket” on Ali the first ones you’ll get should be of the same sort. It’s only unusually wide, but if you know the number of pins and the distance between the rows then you should easily be able to find what you’re looking for. Digikey or wherever might even have filter categories for both, which will give you a part number at the very least.

>> No.1946608

i've ordered some. hopefully its the right size but if not then i'm only out £2.

>> No.1946612
File: 765 KB, 2448x3264, 1590640510423.jpg [View same] [iqdb] [saucenao] [google] [report]

Hey /ohm/,
I have a popular/cheap chinese 3D printer and I've been told that pic related is a fire hazard. It's the input wires to the small logic board from the power supply. They're 24V (probably lots of amps). The wires are tinned and they are pinched by those popular green block terminals.

Why is this a fire hazard? How do I fix it if it is? I've also been told that many people put ferrules on these wires. Why are ferrules any different than these tinned wires?

>> No.1946652

Wasn't on the chins for a while.
Reposting my question:
>My multimeter crapped out finally. I could fix it probably, but I'm at the point I want to just upgrade.
>Anyone have good recommendations for electronics design?

Also in response to the reply I got;
>What price bracket are you looking for? If you want something legit, then a Fluke or similarly reputable brand is the way to go, but they're 8 times more expensive than a reasonable chink meter with similar specs (ANENG AN8009).

I don't mind shelling out a couple hundred for a good multimeter.
The issue I'm seeing when hunting for multimeters is that none of them have "all the features" so to speak. I've seen multimeters that can measure capacitance, inductance, test BJTs and MOSFETS, etc. Don't know if these are just meme features, but to me I believe they may prove useful.

I would like to spend the money and get something that can test a wide variety of components and parameters, but it seems more so that I have to get several multimeters to get every niche feature covered.

>> No.1946665
File: 79 KB, 1000x1000, 10PCS 32Pin Round DIP32 SIP IC Sockets Adaptor Solder Type Gold Plated machined.jpg [View same] [iqdb] [saucenao] [google] [report]

>will these work or no?

yeah. the spring has metal on both sides of the pin, so that's the minimum requirement. but, since you're a klutz, and it's a one-off, you should go for more than the minimum. i.e. the gold standard in sockets: machined sockets. they wont be destroyed by simply plugging the chip in, and pulling it out, a couple of times, like yours was destroyed.

actually, there's one step above machined sockets, and it's ZIF sockets. but one likely wont fit coz they're wider.

>Why is this a fire hazard?

i dont see any. you dont need ferrules, as tinning accomplishes the same purpose: keeping the stands together.
of course, any kind of connector adds resistance to the circuit, so you might have a hot connector if there's a lot of current. to avoid, solder the wires, and maybe use thicker wires, and make sure they're copper, not aluminum or steel. (does a magnet stick to your wires? if yes, you've been chinked)

>> No.1946685

Inductance is a rare feature on a DMM. I’d mainly care about testing in-circuit components with your meter(wide range of voltages and currents, truerms, high voltage safety, ncv sensing, etc.), inductance and capacitance and transistors and such are better suited to an LCR meter or measurement with a scope. If you google “multimeter spreadsheet eevblog” you’ll find a forum post with a rather comprehensive and somewhat recent list of a bunch of mid-upper brands and their different models.

Other notable features include data-logging, min/max/average, smartphone connectivity, connectable amp-clamp, maybe even rechargeable battery. But you should see all that on the spreadsheet’s many columns.
And personally I like autoranging, but some swear against it.

A chinky transistor tester from AliExpress (does LCR, BJT, FET, etc.) is a cheap and surprisingly good tool to have around. Get one with a case, because without one they’re a bit too delicate. Might want to bodge it to use banana sockets too. For accurately measuring inductance and capacitance, it’s best to use a 4-wire tester of some kind, usually a benchtop model. If you don’t need that accuracy, go with the chinkyLCR.
Oh and get a cheap alibay logic analyser while you’re at it, they’re really handy. Can’t say I can say the same thing for chinky scopes though, especially the single-channel fuckos.

>> No.1946708

Has anyone played with ESP8266? What a weird little fucker. A million different versions each has their own quirks. It might be a hobby by itself to figure out how to make them work reliably.

>> No.1946728

You mean differences in specs? Or in things like memory locations and pin IDs and such? If it’s the latter then it should be covered under whatever compiler you’re using, provided you compile it for the right hardware version of ESP.

Never used one myself though, so I’m just speculating. The MCU general isn’t just a gaggle of blinkies on their unos or gaymers trying to make a mech keyboard, there’s quite a lot of people knowledgeable about all sorts of MCU families over there, so I’d recommend that you direct any more specific queries their way.
The MCG/OHM split turned out to be a good thing, in my opinion.

>> No.1946742

I've been wanting to make vector images with my analog scope for some time now. There are three main stages to this that I see. Firstly, converting an image into a vector one, with constant linewidth. I could do variable linewidth to mean intensity or duration on the scope later, but baby steps for now. Secondly, I'd need to generate a continuous path through all of these lines, jumping over to islands as fast as possible or using my scope's blanking input. Then I'd just need to create continuous X-Y voltages from this path using speedy DACs. Along with a simple trigger signal, and any required blanking

Step 1 is the most difficult part, and it's probably something I'd have to do via a proper image editing program like adobe illustrator. Even then, there's still quite a bit of post-processing to do by hand. I'd like to be able to write a program to do this for me, but I'm not exactly confident in my ability to do so. Contrast mapping would be somewhat easy, but actually drawing lines and spitting them out as whatever format is convenient sounds like some nasty 2D numerical equation solving, especially to trace the entire image. Not impossible, but it would take a while to debug and might run somewhat slowly.

Step 2 is something I don't know how to tackle, all depends on the vector format.

Step 3 would just be splitting the whole thing into equal timesteps and extrapolating (straight) lines between the coordinates. And rounding it to suit my DAC's resolution.

Ideally, I'd do all three steps on an MCU and be able to use the thing as a computer display or whatever. Might get around to that in a few years if my knowledge allows me to cram a neural network to do all that for me in the MCU, but for now that's a distant dream. A dream of watching all of SEL in vector format.

Any thoughts? Anyone done this before?
>inb4 those guys and their oscope music
I know, but you have to pay for their software IIRC and it's not exactly what I'm looking for.

>> No.1946761

Thanks anon, I'll look up that eevblog spreadsheet and try figure it out from there. I had a feeling those component test features might have been memish, I will look into those alternate testers. Just don't want to take up a huge footprint on my bench, so I'll have to be mindful of that.

>> No.1946765

>Anyone done this before?

i downloaded some audio files from some place (dont remember where) plugged L+R into my X and Y and watched some very pretty animations. zero work, zero outlay, 100% orgiastic thrills.

>> No.1946774


Yeah the video is called "How to Draw Mushrooms On An Oscilloscope With Sound" on YouTube. He's got a bunch of them where you just hook the left speaker to the X deflection and the right speaker to the Y deflection.

>> No.1946777

Yeah that's the "those guys and their oscope music" I mentioned. Can just buy the song and mock up a 3.5mm to dual BNC adapter and watch that shit forever. But I want something more customisable. If nothing else, a short animation looping on my scope.
The oscope music guys are going the extra mile by making the waveforms all have both musical and visual aspects, which I've no need for. I'm wanting to generate the images from existing images/animation, which may or may not be more difficult.

If I wanted to make raster images, I could just use something like an LM1881N along with supporting circuitry and a composite video converter, but that's not as cool as a vector image.

>> No.1946787

Is KiCads spice simulator on par with stuff like LT Spice?

>> No.1946790

It's all spice under the hood, I doubt there's much difference as to the accuracy of the simulations. Even if KiCAD is missing something, it's very likely that you or someone else could whip up a python addon to improve it. KiCAD has a larger part library than LTSpice, but I'm not sure how many spice models it actually has. I find LTSpice much easier to make quick prototypes in so that's what I use, but if you're iteratively designing a circuit to fabricate and need to simulate it, KiCAD is probably more convenient.

I've never gotten into simulating via KiCAD much though so don't take my word for it.

>> No.1946813
File: 16 KB, 400x221, PIGEON-MAGNET-DECOY-Machine-Rotary-Hunting-Shooting-Bird-_1.jpg [View same] [iqdb] [saucenao] [google] [report]

Got this pigeon magnet i wanna be able to switch on and off from 25 meters away, it's 12 volts, can I just run 25m of cable from it to the battery or do I need some kinda circuit where the battery is next to the magnet and a switch cable is next to me?

>> No.1946830

what in gods name is a pigeon magnet?

>> No.1946838

you put dead pigeons on the end of those rods, they spin and it simulates their landing behaviour, other pigeons love it so they some in to get shot

>> No.1946842
File: 101 KB, 1280x720, wheeeze.jpg [View same] [iqdb] [saucenao] [google] [report]

>pigeon magnet
pic related
you should be able to run a long cable, just make sure it is thick, like 13A mains cable. there's a slight change it might introduce some resistance that could cause you problems but with a longer cable that resistance will be lessened.

>> No.1946843
File: 752 KB, 200x200, bowie laugh worried.gif [View same] [iqdb] [saucenao] [google] [report]

sorry, with a THICKER cable, that resistance will be lessened.

>> No.1946852
File: 27 KB, 459x612, eeerwer.jpg [View same] [iqdb] [saucenao] [google] [report]

fuck sake, i bought the bits for a relay switch

please tell me i've made it even marginally better..

>> No.1946853



>> No.1946862
File: 90 KB, 850x589, lol dogs.jpg [View same] [iqdb] [saucenao] [google] [report]

>please tell me i've made it even marginally better..
yeah you have :) this negates a lot of problems the long wire can cause, defo better.
so now you have a relay controlled by a switch which is still 25 meters away but draws less current and has less problems, lets hope you got a relay that can handle the current your pigeon magnet draws.
pigeon magnet, ffs lmao

>> No.1946863
File: 68 KB, 1242x1121, dog of judgement.jpg [View same] [iqdb] [saucenao] [google] [report]

oh and you could show us a circuit diagram of what you plan. then we can help you make sure you got it right.

>> No.1946888

I asked there, no responses. There are many different versions of this module with different form factors, pinouts, pull-up-down requirements, firmware versions, etc. Arduino forums are full posts wondering about various errors and crashes etc. But after playing with it for a few days, it is not as bad as it sounds. It is just there is a lot of conflicting info since people may be talking about different versions. But most importantly, it is very sensitive to minimum current requirements, a good power supply and good connections so forget about breadboarding it. Once I soldered all connections and figured out some annoying gotchas, it seems to be working fine.

>> No.1946890

Dev boards are manufactured by several factories and vary in layout, materials and quality. Many of them use counterfeit chips which oftentimes don't meet specs of the original, or exceeds tolerances.

>> No.1946891
File: 39 KB, 640x791, daddy.jpg [View same] [iqdb] [saucenao] [google] [report]

>counterfeit ICs myth

>> No.1946895
File: 53 KB, 550x269, spot the fake.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.1946896

does it work as expected or it is just an empty package?

>> No.1946897

>Can you spend a counterfeit $20?

>> No.1946899

Ok niggers Im into fixing vintage audio and have been getting by without a sig gen and a scope using a simple component tester and a fuckload of time wasted in troubleshooting. for a while but I got a tuner that's fucked and this shit is not fixable without a signal generator and a scope. Problem is I dont know what makes a good and fully featured scope or Sgen. Since this is mainly a hobby I would like to keep this thing on the cheap and I'm going for used. What do I need to look out for and how much should I be getting ready to spend for something nice and that wont shit the bed in a year ? Im willing to eventually fix some thing or make some modifications to get something nice. If I have to replace the caps on a tektronics or something and it wont kill the budget I'm ready to get my hands dirty.

>> No.1946901

the cheapest digital scope and signal generator you can find will work for audio.

>> No.1946902

Well I do audio but im buying one to work on RF. I also want something that is good enough to let me do most repair jobs. Im thinking analog or digital video AV equipment and such. I would also want it to be good enough to keep.

>> No.1946904

Its not empty that would require a bit of actual work from chinks. What they do most of the time is the get chips from recycling and try to maximise profits by disguising a chip that has the same package and looks close enough by giving it the model n° of a chips that's worth more.

>> No.1946931
File: 122 KB, 1032x774, wiring.jpg [View same] [iqdb] [saucenao] [google] [report]


here's wot i done

>> No.1946941

i bought a lot of STM32 micros on aliexpress never had any problems. All genuine parts. And how the hell can you fake something as complicated as STM32F4 or F7.

>> No.1946944

Does that relay switch properly with 9V on that length of line? If not you will need to up the voltage.

>> No.1946949

indian here, uncle does shady business with chips and refurbs a lot of microcontrollers from discarded consoomer products, but the markings fade/disappear in the cleaning process, that's why they look so sketchy. im glad that i don't live there (India) anymore (i moved when i was <1 yo)

>> No.1946951

it makes less noise switching compared to when i had a short cable on it, but it still works

>> No.1946952
File: 15 KB, 963x275, audio probe.gif [View same] [iqdb] [saucenao] [google] [report]

a 555 oscillator and a audio probe, duh

>> No.1946956

It might not be engaging very solidly and as that battery wears down it may quit working all together. That looks like a standard automotive relay, so 12-14 volt normal operating range, I would jump the voltage up to 16 to account for line loss.

Also your signal line is way thicker than it needs to be, the current to switch one of those relays is tiny (makes no different if that's what you have lying around just overkill). Conversely the line going to your magnet looks too small if that thing is actually drawing anywhere near 30A.

>> No.1946957
File: 25 KB, 465x385, bad booster.png [View same] [iqdb] [saucenao] [google] [report]

i am building a FUZZ pedal for guitar and i want to double the output volume with the least amount of parts possible. the signal coems out of an previous gain stage biassed at Vcc/2 and i'd prefer to not use an opamp, would this work?? (im not home to test it out this weekend/monday)

>> No.1946967

My amplifier knowledge is lacking but what's going on with the 2 diodes here? One seems pointless and the other seems like it's going to block half your 40Hz AC waveform. Other than that it's a common emitter amplifier, provided your math is right it should work. No idea if you will get the sound you want from it.

>> No.1946970

the signals sauce has a +4V offset so the diodes will be forward biassed at all times., this is to set the bias point properly, so the signal does not prematurely c̶u̶m̶ clip. the diodes are LED in the simulation software to get to the right bias point, but more 4148's can replace them to get the same bias voltage

>> No.1946977

lmao what kind of margins is it even possible to get out of that?

>> No.1946979
File: 5 KB, 759x158, square wave.png [View same] [iqdb] [saucenao] [google] [report]

instrument lever is generally less than 1V pk-pk

>> No.1946994

The problem with the ESP8266 is that they're not designed to be general-purpose MCUs in the first place, I think a majority of the documentation we have on them was reverse engineered. That and >>1946890. So to ensure you end up with the least amount of hassle, buy a bunch of bulk ESP8266s from somewhere slightly more reputable like LCSC, so you only have one type to be working with. Not sure if you can configure your compiler with those pinouts and such, but writing them down on a text document should be sufficient.
Isn't the point of an MCU that you upload your own firmware to it in the first place? Or are you trying to program it via some sort of bootloader?

For scopes, you have quite a few options. Personally I'd recommend a 2nd hand analog scope, should be able to get one with decent specs for <$80, depends on specs. Dual channel is a must. You could also get a cheap USB oscilloscope, probably fine for what you're doing with it so long as you don't have any ground loops to your computer. Might not be very convenient if your only computer is a desktop in a different room though, and I'm unfamiliar with their required computer hardware or software. There are also the occasional 2nd hand scopemeters you may find, which depending on their specs could be a handy choice. Then there's the "just use my sound card: choice, which consists of a 3.5mm audio cable with a bit of supporting electronics for overvoltage protection and input impedance insurance and such. Sometimes with opto-isolation too. Never done it myself, but there's probably software for it.

For fgens, your cellphone with an audio jack plugged in should be sufficient for any audio work. Just play an mp3 with your desired test signal, perhaps with a variable-gain op-amp circuit. Plenty of other options, but whatever you do ensure it can put out a triangle or sine wave; squares aren't good for audio testing.

What frequency?

>> No.1947004

all you need are the masks, the gooks doing the clone don't care how it works, nor do they need to care.

>> No.1947007
File: 103 KB, 1000x800, wiring for dollars.jpg [View same] [iqdb] [saucenao] [google] [report]


that's no good. you need to run the relay from the same 12V battery coz your 9V is gonna die very quickly.


you're doing it very wrong. just bias your transistor in the usual way, then put a cap in series with the input which will remove the DC component.

>> No.1947009

yeah anon, I am sure people who run IC foundries in their basements are pretty dumb

>> No.1947016

This doesn't work anon. Your toggle switch is hooked to the batt + and - on the relay switch circuit. When you close the toggle switch you are going to have 12V on both sides of the relay switch circuit and the relay will do nothing.

>> No.1947041

Thats a neat trick but that wont help me fix a tuner.
I've been told 50Mhz is about as much as il'll be expected to need a scope to handle.
However to adjust a tuner I need a generator that can cover the whole FM range at the very least so I can calibrate a tuner. Best would be something that allows me to fix a crt and I don't know up to what frequency these things can opererate.
I don't have a sound card and I would rather spend the money on an actual scope.
80$ is about what I was aiming for the scope.
What should I watch out for when buying these ? Any defects or telltale signs something is fucked? How about probes ?

>> No.1947048
File: 100 KB, 1000x779, wiring for pennies.jpg [View same] [iqdb] [saucenao] [google] [report]


right you are, nigger.

>> No.1947049

Your requirements of Frequency range drive your Scope choices, not the other way around. If you are dealing with audio just buy whatever, if you are dealing with <50Mhz just buy whatever, if you are dealing above 100 Mhz then we can discuss.

>> No.1947050

CRT screens themselves are slow enough that an analog scope can run on its signals just fine. At least, if you get one with TV-H and/or TV-V options, which digital scopes are less likely to have. If you want to tune into an analog TV broadcast, then A: that's kinda old tech that might not be in use anymore, and B: it would require a rather specialised piece of kit that I'm not qualified to tell you about. It would need pretty high oscillator stability, necessitating that it's running with some sort of crystal. IIRC the ham guys have one or two favoured DDS ICs, might be able to ask that general for advice.

As for selecting the scope itself, browse about on whatever local 2nd hand website is most popular and hunt for a good deal. You want 2-channel, probably at least 50MHz, and nothing particularly ancient looking. Old ones look cool, but are likely way out of calibration. I got my scope almost freshly calibrated, the university was getting rid of them in the first place because of the continual calibration cost.

>> No.1947069

Sorry when I mean I've been told I can get away with 50mhz its with my use case taken into account: Ie mostly audio equipement and tuners.
About TV's I dont partcularly care about TV broadcasts more Video input signals.
I've considered picking up my Uni's stuff but thy got rid of their stuff a while ago.

>> No.1947074
File: 8 KB, 644x603, Screenshot_20201108_140337.png [View same] [iqdb] [saucenao] [google] [report]

try this mane
>Rc/Re gives gain of 2
>100k (86k in theory) gives base current
>10u cap input

>> No.1947079

I have an old circular saw that spontaneously died. What typically makes them die?

>> No.1947081

I've been thinking maybe I should spend a bit more and get something a bit nice I can keep for a while and have spare parts for.
Would something like a Tektronix TDS 420 at 160 euro be a good deal ? No probes but looks to be in nice condition.

>> No.1947141
File: 4 KB, 220x189, 220px-1-bit_full-adder.svg.png [View same] [iqdb] [saucenao] [google] [report]

Why is this called a one-bit adder? It has 2 inputs (A,B) and 2 outputs (S, C_out) so shouldn't it be a 2-bit adder? Or is it called 1 bit because 2^N when N=1 is 2?

>> No.1947145

>150MHz @ $160
yes that is a good deal

>> No.1947148

Guy sells a matching tektronix AWG 2020 SigGen at 10 bucks less. One channel fails check (error 642) and the BNC for CH2 is gone. Both devices are known to have shit caps all over that make them shit the bed. Im going to try and haggle the guy because they will shit the bed in a month if this is not dealt with + some rebate for taking both. I dont expect to pay more than 25-30$ for the recap + chink BNC. Good deal or nah?

>> No.1947149
File: 662 KB, 1112x845, eagleboard.ga_5860.png [View same] [iqdb] [saucenao] [google] [report]

>he doesn't charge his li-ion batteries to 4.8V for extra capacity

>> No.1947163

the schematic i posted earlier works. i have a green LED in the diode position, this gives the right bias
i want to minimise capacitor count as much as possible because of a challenged issued by a friend of mine.

>> No.1947172

>Isn't the point of an MCU that you upload your own firmware to it in the first place?
Yes, I'd say it is the most common approach by far.

>Or are you trying to program it via some sort of bootloader?
I am trying to use the stock firmware running in the "AT mode", basically like a modem, over a serial port. The AT command set mimics their native API and provides a similar common functionality: establish connection, send/receive data etc. I don't want to mess with their SDK and compile the code for their exotic platform since I have zero interest in learning it. I mostly work with STM32 which I know well. So I prefer to write code for STM32 and access the ESP8266 as a peripheral over a serial connection. This is not a very popular approach since ESP8266 became extremely popular in the Arduino community after they added support for it via a wrapper around the ESP8266 native SDK. So it is relatively easy to program and flash in from within the Arduino environment using the same tools and code, but instead of targeting Arduino you target ESP8266. But I don't do Arduino either and all my attempts to maybe consider that failed miserably since every time I see the word "sketch" I cringe so hard and give up.

>> No.1947189
File: 8 KB, 751x825, wfsdf.png [View same] [iqdb] [saucenao] [google] [report]

so this is wrong?

>> No.1947191
File: 53 KB, 595x398, 1509600162105.jpg [View same] [iqdb] [saucenao] [google] [report]

i kind of see what you're doing there with the diodes, but it might be better to pass the signal through a cap, remove the bias entirely, then set the bias of your transistor with a voltage divider.
oh like this >>1947074
aashsfds wish you'd said sooner.

>> No.1947192

Brushes, water or other ingress, perhaps a cracked gear or sheared shaft or retaining pin, could be anything. Take it apart if you want to know.

When you chain them together to add two 8 bit numbers, you need 8 of them; one for each bit.

If you’re happy with the price, go for it. Recapping shouldn’t be too much of a hassle, and chink BNCs are fine in my limited experience.

I’d rather see a graph going down to 4V or lower. Could measure the lifetime use with total watt-hours expended for a more even comparison.

Oh, you’re using it as it was originally intended. Never crossed my mind for some reason. I can only suggest that you either find some other wrapper or compiler for ESP coding, perhaps editing an existing compiler to work better with your programming method and ESP hardware model, or get more official or at least more consistent hardware in the first place.

>> No.1947196

mate of mine had a circular saw that died. the commutator exploded.

>> No.1947200
File: 51 KB, 474x711, harold diy 2.jpg [View same] [iqdb] [saucenao] [google] [report]

different anon, no, it's not wrong, you can do it that way.
there's just a chance, as the other anon said, that you may not have enough juice for your relay now.

>> No.1947287
File: 215 KB, 754x1161, power supply jpg.jpg [View same] [iqdb] [saucenao] [google] [report]

How can I determine the input voltage on this power supply? On the heatsink there is a LT1170 and a BYW81P-200. I assume input is 230v AC based on where the item is from, but I also afraid that I can destroy it if its wrong.

>> No.1947291

>but I also afraid that I can destroy it if its wrong.
looking at those caps it's not long for this world anyways

>> No.1947295

what are the caps closest to the green connector rated for?
There's also a fuse rated 6.3A under the upper fuse holder.

You can tell by the isolation marks and the XY or X or Y cap near the connector that that's a section rated for some kind of mains.

The board is made by a German company.

I'd say if the caps are rated for whatever the P-P voltage of 230v RMS is, then yeah, it's 230.

>> No.1947297

Thanks guess that clears it up. The caps are rated 50volt so now I'm thinking it's a 24-12v dcdc supply. 24 volt DC would make sense since its from a remote control for a crane on a ship.

>> No.1947320

Oh the most likely part to break is often the switch itself, since it sees a lot of cycles. Also if there’s a suppression capacitor across the motor, it might have failed short and blown a fuse or switching transistor. If it’s variable speed, then it’s using a MOSFET (or TRIAC if AC), which can definitely overheat and break if cooling is insufficient.

If you plan on fixing it, check and fix the most solid parts like the gears or commutator first, with any silicon shit getting fixed last. That way you don’t kill your transistors a second time.

>> No.1947332

Got it apart, everything mechanical looks fine. The previous owner said they saw a spark inside the motor when it died. I haven't done any electrical tests on it yet, so I don't know for sure if the motor's fried or it's a driver problem. Couldn't I wire the motor brushes to a DC bench power supply to see if the motor is fried?

>> No.1947335

>a spark inside the motor when it died
That's normal for a brushed motor even when it's fine. Take the brushes out and see if they're intact and long enough first. Brush removal should be relatively simple, as it's usually intended for the customer to do so a couple of times in the lifetime of the tool.

Is it AC or batter powered?

>> No.1947340

AC powered.
The brushes seemed fine, they were curved with the commutator and had a lot of material left.

>> No.1947365

Check where the wires meet the comm-bars, that's a common break point that could make a spark when going open-circuit. It's why more modern tools often have them covered with epoxy.

>> No.1947369

>Couldn't I wire the motor brushes to a DC bench power supply to see if the motor is fried?

it would tell you if one or more coils are open, but so would an ohmmeter, which is a lot more gentle -- like Biden kissing his hot loli niece. but it's not recommended unless the supply has current limiting, or you might burn out a coil. and to make it run, you'd need to power both rotor and stator, in series, at a fairly high voltage, maybe 30, 40, 50V volts, and lots of amps.

>> No.1947373

>I'm thinking it's a 24-12v dcdc supply.

probably, but there's a bit of weirdness with the fuses. if you were stepping down by 2x, you'd expect the input fuse to about 2x smaller, but it's 6x bigger, which would suggest you're stepping up.
anyway, the caps always tell the story.

>> No.1947511

if the other ways better i'll do that,

is a battery just activating a relay bad?

>> No.1947560

how do i generate high voltages to accelerate a crt electron gun but also make sure not to kill myself
since i only need high voltage plates to accelerate electrons, no actual current flows though right
that makes me wonder though, where does the electron kinetic energy come from

>> No.1947561

electrons gain speed every time someone who does not know how potentials work try to deal with them.

>> No.1947564

i was thinking the electrons appear magically but now that i think about it, the heating element brings them close to the plates so the energy comes from it

also what happens once the electrons hit the phosphor, the negative charges have got to go somewhere
is the phosphor conductive and grounded?

>> No.1947565

The phosphor coating is positively charged with HV from the flyback transformer. The electrons are negatively charged and they're steered by the yoke coils.

>> No.1947569

How can MOSFETs allow a drain-source voltage lesser than the gate voltage? BJTs can't do that, can they?

>> No.1947570
File: 11 KB, 280x286, degauss & ground.jpg [View same] [iqdb] [saucenao] [google] [report]

>is the phosphor conductive and grounded?
There is a ground strap around the tube.

>> No.1947572
File: 117 KB, 812x800, sneeef.jpg [View same] [iqdb] [saucenao] [google] [report]

I built a boost converter from random parts I had laying around, it takes in 5V from a power bank and boosts it upto 70V.
Now I know in theory "how" boost converters work, but I have absolutely no idea how to design a proper one. I don't know how to calculate the switching frequency, or duty cycle to maintain a particular voltage or anything. Where can I learn about these arcane arts? I tried running some ngspice simulations in kicad but they take ages and unless the time base is extremely small, it gives me negligible information about the waveform that I can expect.

>> No.1947573

Look up datasheets for boost converter ICs and study the example circuits.

>> No.1947581

I'm current not using a boost IC or anything, just a MOSFET, with a random unmarked wounded inductor, and an atmega8 acting as the gate driver
This is a purely an academic exercise, I'm not planning on using this anywhere

>> No.1947587

Then look up the datasheet for the MOSFET and compare designs that use the same component.

>> No.1947598

how about u use a signal generator or 555 timer and a strong low pass filter, put like 10 volt on the input and just see what comes out., then just crunch some digits and .... booom... you know it

>> No.1947645

parameters usually are
>power rating
>nominal output voltage
>input voltage

>> No.1947661

any recommended electronic videos or playlists for absolute beginners?

>> No.1947675
File: 168 KB, 490x1096, Sz8lv6O.jpg [View same] [iqdb] [saucenao] [google] [report]

Any of you fine gentlemen ever went dumpster diving for electronics and found something?

I always see ppl like EEVBlog go dumpster diving and I´m amazed what he pulls out of the bin.

>> No.1947684

Yeah, small transformers, micro controllers, IGBTs, small motors and belts..

>> No.1947686
File: 26 KB, 600x497, cell1-600x497.jpg [View same] [iqdb] [saucenao] [google] [report]

I am a brainlet trying to understand solar power.
Lets say I have a solar panel connected to buck converter and a variable resistance. If I do not care about the voltage in the load (just the power) how do I select the duty cycle for different resistors? I can grasp at how a converter will make the panel achieve MPP with different loads.
>imagine pic related is 0-60v instead of 0.6v
>how do I select the duty cycle for lets say, 1k ohm and 100 ohm?

>> No.1947690

>is a battery just activating a relay bad?

it's the quality of the battery that's bad. it's 9V, and the most common type of relay is 12V. also, a 9V battery is fine for a radio where it'll draw 15 to 20mA, but a relay can take 10x as much current, so the battery will die pretty quickly. and it's not rechargeable, like the 12V, so that's gonna get expensive.

>> No.1947693

alright I just assumed the converter IS operating at the MPP, used that power value to get the voltage at the load to get that same power.

>> No.1947716
File: 71 KB, 917x725, display.png [View same] [iqdb] [saucenao] [google] [report]

I'm trying to control the backlight of an LCD with a pin, does the transistor need to be there or can I just use the pin and a resistor? Please no bully, I'm new. Thanks in advance!

>> No.1947719

>NPN on high side
As for driving from MCU directly - depends on forward voltage drop of backlight LEDs but even assuming it's about 3V - it would be about 8-10mA current draw which is not advised for MCU pins

>> No.1947720
File: 710 KB, 2048x1536, Photo0529.jpg [View same] [iqdb] [saucenao] [google] [report]

all my computers
I have 3 more, one of which i am currently posting from right now.
lots of speakers, a full sound system, so much cat5 and cat6 cable and electrical cable, i found fucking at least 300 feet of it in one skip once when i was about 19, i am now 31 and i still use it and have not run out.

>> No.1947730

>unemployed IT grad
>only knowledge of electronics is from making Arduino projects and fucking around with kicad, made a PCB before
>make an account on upwork and work as a "freelance" embedded engineer
>people actually come to me, do relatively simple jobs designing circuits and PCBs
>learn a lot eventually
>very polite older brit guy comes to me, he's getting into Arduino
>wants me to design a battery powered ESP32 board with many different sensors etc
>will be good practice for me so agree to do it for 70$ in 7 days
>finish job, guy tips me 10$
>ff now, he gets the board manufactured
>board doesn't work
>approaches me for help

What do lads? I don't generally post here but I really feel bad for the guy, he was so polite too, unlike most fags on upwork

>> No.1947734

>board doesn't work
What exactly doesn't work? No voltages at all?
ESP32 cannot be programmed? Some sensors do not work?

>> No.1947736

Fix the problem or refund him. How is that unclear?

>> No.1947738

Thanks for this. What's the best way to drive the LCD backlight with a pin then?

>> No.1947751
File: 80 KB, 917x725, 1604949731657.png [View same] [iqdb] [saucenao] [google] [report]

For NPN transistors (and PNP) always use resistor between base pin and MCU pin (or other) because otherwise MCU will sink as much as it can from base to emitter and blow it's brains out.

Using NPN on high side also has some other issues but a bit too technical for now (if you want NPN on high side - use optotransistor)
And recommend avoiding PNP transistors for time being until you learn more

>> No.1947758

thank you very much for this. I'm very intrigued about all of this information. Why is highside bad exactly? Thank you very much once again

>> No.1947776

NPN on lowside is idiot-proof design and very simple

Using NPN on high-side for driving LEDs you have voltage drop of transistor Vbe (~0.7V) and LED forward voltage drop (~3V for white) so you need about 3.7V on transistor base pin to even begin turning on transistor. And even if it turns on at such control voltage it will be very dim, might be better if BLU_PIN can provide 5V but not nearly as bright as when NPN is low-side switch. For fun I recommend building both circuits and test them.

>> No.1947852

Sounds like you designed an open-loop converter. The simplest way to control a boost converter is to put the output voltage through a voltage divider, and feed it into a comparator alongside a reference voltage (I like the TL431). Then use that comparator to turn off the oscillator whenever the divided output voltage is higher than the reference. This causes somewhat bad ripple though, most proper switching controllers will lower the duty cycle as the voltage gets higher. Instead of a comparator, they’ll use an “error amplifier”, a differential amplifier of some sort with a fixed gain, where the output is used to vary the pulse width. If the gain of this amplifier is too low, then you’ll get poorer output regulation as the voltage will dip under load. If the gain of the amplifier is too high, then the circuit may become unstable. Course, they also have an oscillator topology that can have its pulse width varied by a control voltage, a push-pull MOSFET driver if not an integrated FET, thermal monitoring, maybe some sort of bootstrapping circuit.

>> No.1947881

For simulation software, does one really need to need resistors in something like logisim?

>> No.1947900
File: 4 KB, 464x241, high-z.png [View same] [iqdb] [saucenao] [google] [report]

resistors in logisim are used to pull nets that are otherwise in a high-Z state (they will be blue-coloured) to a well-defined value (0 or 1).

This is important, since the high-Z logic will propogate as an error when used as an input to logic gates. see pic.

>> No.1947929

Not him, but:
Does it support logic that requires external pull-ups? Like an and-gate that only pulls its output down via an open-collector? And only pulls its input up via an open emitter? Pretty sure that’s what 7400 logic does, or at least did. Not strictly high-Z, but not a solid output either. Let’s you do things like communing two outputs for a logical OR. Or maybe it’s NAND.

>> No.1947993

Voltages seem fine, it's the ESP. He cannot program it using the onboard USB and CP2102. The ESP doesn't send anything over serial for some reason

>> No.1948050

>not sending even boot messages
just post full schematic

>> No.1948054

>using a USB-UART bridge
>not having an ICSP header as well
pleb design

>> No.1948055
File: 118 KB, 1280x720, eevblog 54622d.jpg [View same] [iqdb] [saucenao] [google] [report]

Local auction site has a loaded (all probes, docs, etc) HP 54622D up for under $150USD. 100MHz 2 analog channel plus 16 digital channels. I know only 2 analog channels can be slightly painful but other MSO scopes are significantly more expensive for just the system unit with no probes and stuff. Should I go for it?

>> No.1948095

How easy is it to get data out of it? I hope there's an option other than the floppy drive. If not, maybe there's some SDcard-floppy equivalent to those audio cassette-aux things. If you're not going to be exporting data, then check to see what math functions it has. Like an FFT or whatever.

It doesn't look like a bad scope all things considered, though you can get 4-channel modern bottom-end Rigols for $300 or maybe less. If you're happy with that price, then it looks like a fun scope to own.

>> No.1948204

what happens if you reverse a bjt(collector and emitter the wrong way)
does it just reduce gain or something or is possible for it to die instantly

>> No.1948228

If it has a floppy header USB adapters are cheap.

>> No.1948239

thats called reverse biasing . it can create signal noise and stops using the base all together if its a npn. this can be used in cases where you want to create some kind of pulses. it does not work with pnps only npns. why? npns because of the sandwich can go super conductive when reverse biased where as pnps cannot.

it will kill your bjt and depending on your input current it may do it slowly. it will create pulses but it will wear down the part faster than normal usage

>> No.1948240

i guess extended explanation for why pnps wont do it would be the size of the center slab of the 3 part sandwich. the wrong doping will not permit it to do it. you just kill it little by little. funny thing is on some normal diodes you can do it to

>> No.1948248

Transistors can be reverse biased on top of the individual pn junctions being reverse biased?

>> No.1948260

yes . its something people were showing off a few years ago. a c945 (npn) can do it and ring (as well as tons of others) with a capacitor it makes a relaxation oscilator. in a circuit it acts like a zener or spark gap with a minimum voltage to cross and the capacitor stalls out that process. but the ones that ring give a set frequency with their noise. oddly the youtube vid is identified as spam. but ion youtube many had done it. google single bjt oscilator reverse biased

>> No.1948277

I want to re-wind an alternator so that it outputs 48 volts DC instead of the standard 12, so as to improve efficiency

readily available car batteries are 12 volts though.
I'm about to sound completely retarded, but if I put 4 12 volt car batteries in series (=48 volts), would I then be able to charge those batteries with the 48 volt alternator?

>> No.1948286

Alternators need to run at high RPMs at all times to be efficient, and you need a charge controller for multiple batteries.

>> No.1948287

Also, car batteries are 13.6V at full charge. If your battery dips to 12V it's toast.

>> No.1948288

it's more of an experiment if anything. I have a spare car alternator, my friend has a shitty little bicycle -> moped engine he isn't going to use, and I wanted to see what I could redneck engineer out of it

>> No.1948311

What is wrong with 12v?

>> No.1948320
File: 1.62 MB, 3200x2400, IMG_20201110_174545.jpg [View same] [iqdb] [saucenao] [google] [report]

hello /ohm/, I have a cursory understanding of electronics and I am a decent solderer (have done loads of repairs on my dad's racing motorcycles after crashes).

I am currently taking care of a mentally handicapped individual, and I'd like to lower the max volume their device can go to.
The volume control is digital, so not a knob you can just glue shut, so I said I might know a solution.

Attached is a photo on the inside, it is my understanding that if I solder resistors in line with the cable and the connector for the speaker, it will lower the volume.

Am I correct in that assumption? How do I determine what kind of resistor I need, trial and error?
Can I use a multimeter to measure the speaker's resistance and make an educated guess that if I put something around 20% of that in ohms infront of it, it'll reduce the volume by 20%?
Is it safe to buy resistors from aliexpress?
Would heating up of the resistor not be an issue?

Reduced battery life is not something I'm worried about, but if there's a more elegant solution I'd love ot hear it.

>> No.1948325

somewhere should be amplifier for audio, see what IC it is and maybe it has adjustable gain so you could replace some passive components and reduce max volume.
Now that I wrote that - if there is a volume knob see if it is digital or controls amplifier gain directly then just solder resistor in series (or form voltage divider) to reduce max volume

>> No.1948327

it wastes a shitload of power as heat within the windings since you need a higher current to get the same amount of power
at a higher voltage, you waste a lot less

>> No.1948328

Nah it's 2 push buttons on the front, one up, one down.
As for all those other things you said, they're too technical for me I imagine because I don't know what you mean.

>> No.1948330

EE grad here with an entry level job. Question time, a more experienced engineer told me almost everyone hates P channel / PNP transistors.

How true is this?

>> No.1948336 [DELETED] 
File: 61 KB, 933x916, zooom.jpg [View same] [iqdb] [saucenao] [google] [report]

Assume a constant voltage input, calculate the series equivalent resistance. (Rseries+Rspeaker)
Calculate the current that will flow:
I = V/Rtotal
Use the formula for power
>P = I*Rspeaker
You will see it changes quadratically with the series resistor size.
>5v -> 8 Ohm speaker
>power = 3.15 W
> 5V - 8 Ohm resistor - 8 Ohm speaker
>power = 3.15W/4 = 0.78 W
>5V - 16 Ohms - 8 Ohm speaker
>P ≃ 3.15/9 = 0.35W
You use what you have to use, they are not used for very high power stuff or super fast things. For the same size they offer usually higher resistance.

>> No.1948338

Also how loud stuff sounds is usually related to the log of power.
>10x more power = 2x more loud
And the speaker resistance is usually written on the back. (8 or 16 ohms usually)

>> No.1948363

Thanks lads, I think I'll add a 2 ohm resistor since it's an 8 ohm speaker already.

>> No.1948378

You need to find the line level signals for R+ and L+ and put a stereo pot on those traces before the amplifier. Set it at 50%, turn the digital volume to 100%, and adjust the maximum volume on the pot.

>> No.1948379
File: 75 KB, 1004x588, 20201110_131411.jpg [View same] [iqdb] [saucenao] [google] [report]

Here's an interesting one if you used a 12 hour clock face what numbers would be at the at the yellow dots?

>> No.1948381

just make sure it can handle the power, a 1/4W might not do it.

>> No.1948386

>2n2222's have 300MHz bandwidth
>any signal above ~10MHz gets sucked up by the junction capacitance to ground/Vcc
Fuck you RF, fuck you datasheets, and most of all fuck you picofarads

>> No.1948387

what about 5W?

>> No.1948388

Those are chunky resistors.They'll work but I was thinking more on the lines of 1/2W lmao. But if it works it works.

>> No.1948390

that's what I see sold online, electronics stores that sell resistors I haven't seen since the 90s.
tomorrow I'll kill an old access point and try to read it

>> No.1948391

By read it I mean "read the ohm values from the resistors inside there", I'm sleepy.

>> No.1948394

Uh you can find 1/2 W resistors anywhere.. You can also just put two 4ish ohms in parallel to handle 1/2W.

>> No.1948418 [DELETED] 
File: 28 KB, 506x355, 50-ohm high power wirewound pot.jpg [View same] [iqdb] [saucenao] [google] [report]

>if I solder resistors in line with the cable and the connector for the speaker, it will lower the volume.

that's right. it's marked 8 ohms, so you wanna use something in the 4-20 ohms region. or get a 50-ohm 2 watt wire-wound pot so you can set it exactly.

>> No.1948423
File: 28 KB, 506x355, 50-ohm high power wirewound pot.jpg [View same] [iqdb] [saucenao] [google] [report]

>if I solder resistors in line with the cable and the connector for the speaker, it will lower the volume.

that's right. it's marked 8 ohms, so you wanna use something in the 4-20 ohms region. or get a 50-ohm 2 watt wire-wound pot so you can set it exactly.

protip: cut 1 wire in the middle, dont unsolder at the speaker, coz if you're a noob, you can unsolder the airborne wire that goes to the coil.

>> No.1948425

Get some RF transistors lad. Right tool for the job

>> No.1948430

Is there a place to get Ai-Net/Unilink connectors? I see that they all use the same snap-together connector, but I can't find just the connector.

>> No.1948514

Sorry I don't know the answer, but how are you liking your entry level job anon? And what are you doing in it?

>> No.1948535

Incorrect. A motor would for 48V would have four times as many windings, hence it would use wire with a quarter of the cross-sectional-area. The net result is that you get the same percentage of energy wasted through the winding resistance. If you don’t believe me, you can do the mathematics yourself.
The only thing you will improve is the losses from transistors, diodes, wiring (assuming you don’t also change the wiring gauge), that sort of thing. Such losses typically aren’t too big in the big picture, but you can do the calculations on those too. There are solar charging ideal MOSFET diodes that might be less hassle and cheaper than rewinding, in case diodes are what you’re worried about.

P-channel MOSFETs are objectively worse than N-channel ones. Just due to how the dopants happen to work. The three main quality markers of a MOSFET (Qgate, RDSon, Vmax), when taken together as some quality value, is maybe 10% worse for PFETs. H-bridges are frequently used with funky high-side drivers so they can only use NFETs.
That said, the same doesn’t really apply to PNPs. Back in the days they were really hard to fabricate in ICs, but now that’s not an issue.

>> No.1948552

Just started electrical engineering. This shit hard holy shit

>> No.1948563
File: 10 KB, 226x223, jew gold.jpg [View same] [iqdb] [saucenao] [google] [report]


if engineering is hard, you should transfer to finance. i did both, and finance is just basic common sense: if they didnt teach you stuff, you coulda figured it out on your own.

plus you'll make twice as much in salary and bonuses. everyone in finance gets big year-end bonuses, even when you go bankrupt.
''Banks Paid $32.6 Billion in Bonuses Amid U.S. Bailout (Bloomberg) July 30, 2009''

>> No.1948567
File: 28 KB, 386x350, mu x.png [View same] [iqdb] [saucenao] [google] [report]

How come for this 4-1 multiplexer, putting in input of (S1,S0)=(00) would result in W? I understand that 00 = the first option which would be W but based on the schematic, shouldn't there be an inverter connecting S0 to the top multiplexer or else both of them would turn off if the input to S0 was 0?

>> No.1948577

>would turn off if the input to S0 was 0?

what are you talking about, dude? there is no turning off or on involved anywhere here. S0 and S1 are selectors, they select of the the 2 possible inputs and carry it to the output.

so if S0 is low, the two (left) selected outputs are W and Y.
then with S1=0 the selected (right) output is W.
it's all obvious and correct.

>> No.1948584

god damn ltspice is much less of a pain in the ass once you discover you can name nodes and have them connect wherever

>> No.1948585

Oh, my brain shortcircuited, thanks bro.

>> No.1948592


>> No.1948618

ltspice + easy to follow visualizations a la falstad for my zoomer brain when

>> No.1948619

I'm building a "midi" controller (not really midi, and not for controlling music, but you get the idea) and I need some recommendations for components. I want some really smooth dials that have infinite rotation on them, what should i go for?

>> No.1948620

something something hall sensor (probably (maybe))

>> No.1948621

component #?

>> No.1948622

rotary encoder

>> No.1948624


>> No.1948625


>> No.1948629


>> No.1948645

Go to digikey or similar website. Go to the category “rotary encoder”. Filter by magnetic or hall-effect, also maybe optical. Then sort by price and/or availability, and check out some datasheets.

>> No.1948664
File: 82 KB, 2519x606, file.png [View same] [iqdb] [saucenao] [google] [report]

where to filter to magnetic/hall effect? searching returns 0

>> No.1948669

Select “continuous”, which should give you what you want, hall or not. If the Hall effect ones aren’t there, they might be under hall-effect sensors themselves, not rotary encoders.

>> No.1948678 [DELETED] 


you're in the wrong section, nigger. rotary switches are very diff from rotary encoders.

>> No.1948682
File: 6 KB, 273x450, encoder type.gif [View same] [iqdb] [saucenao] [google] [report]


you're in the wrong section, nigger. rotary switches are very diff from rotary encoders.

>> No.1948716

Guys, newb /ohm/ here. I'm used to fucking around with code, but decided to intermingle it with physical.

I have an AI accelerator, a google coral dev board which I'm outputting with serial UART with the aim of connecting it to the RX of an arduino board. I bought a Hantek 1080C (digital oscilloscope) to play around with. If I output "hello world" with the board, I can catch it the binary for a brief millisecond on the oscilloscope. But how can I record this to make sense of it?

>> No.1948723


depends on scope. look for something like ''singleshot'' trigger mode in manual.

alternately use a camera with high-speed recording.

>> No.1948725

Would it make sense to put the "hello world" on repeat and then try and and catch it?

>> No.1948728

There definitely should be some sort of single capture/single trigger mode like the other anon mentioned. You can either trigger it off the data waveform itself, or use the CTS/RTS/etc. line from the UART. Assuming it has those lines and that your chinkscope has an external trigger output, that is.

What software are you using? Something like Pulseview might work with it, I know it's good for my shitty logic analyser. Speaking of which, this is the sort of work that could easily be done with a $8 USB logic analyser instead of a scope that cost 10 times as much.

You can do that, so long as you've got a trigger holdoff function so you always trigger at the same place. Adding a short delay between each transmission will help with this. If not it will look a bit of a mess, but still enough to hit the pause button.
Looking at repeating signals is basically what all analog scopes back in the day were good for anyhow, with exception given to those scopes with slow-decay phosphors.

>> No.1948746

Looking for a PWM controller with integrated gate driver. Must be THT. SMD will not work for prototyping, the high gate drive currents will incur a lot of additional losses through the tiny pins on a breakout board.

Just being a PWM controller with totem pole output (eg. TL598) is not good enough it either needs to have one of two things: a.) Two separate power input pins (typically notated as VCC and VC on most schematics) for each totem pole output so that it can be bootstrapped and drive a high side N-channel FET OR b.) it must have level shift circuitry inside the IC so it can drive a high side N-channel FET.

I am considering a TL598 or similar IC + IRS2101 gate driver which together should do what I want but if there is an IC with both functions combined into the same package at a reasonable price that would be ideal.

>> No.1948773
File: 7 KB, 400x400, a t.png [View same] [iqdb] [saucenao] [google] [report]

>Si5351 breakout board
>two 74HC164s as clock dividers
with this, i shall create many octaves of musical notes
>two rotary encoders
>four seven segment displays
>some SRAM
with this, i shall create a preprogrammed sequencer
>pic related
>F sharp as the 3rd note
>encoders for adjusting note and position respectively
>encoder buttons for increasing and decreasing total number of notes
can't adjust it while it's playing, but it's not made for live performances anyhow

>> No.1948774
File: 470 KB, 1000x498, waifuduino.jpg [View same] [iqdb] [saucenao] [google] [report]

For those of you not lurking /g/csg:

>> No.1948776

how can i get full color silkscreen? are there any cheap boardhouses that do this?

>> No.1948778

arduino-chan is best girl!

>> No.1948779

skipped, if you want to do any analog stuff. i personally find that two channels is fine for whatever, but with a sample rate like that you definitely want to stick to digital anyway so you'll probably end up using the analyzer pods anyway

>> No.1948781

that scope costs more than my car

>> No.1948783

not sure what you're asking.
based, but also seems overkill. iirc you only need to count cycles of a 44khz clock to accurately recreate audio; and since actual musical note fundamentals usually top out at ~4khz, you could get away with a ~10k reference clock, or a 32khz quartz oscillator for RTC's
I have also noticed that the Si5351 is slow to update from the arduino, takes like 1sec

>> No.1948788

I'm not using the Si5351 for its maximum frequency, I'm using it for its frequency resolution. Being able to get an A spot-on is of utmost importance to me. If I order 5 of these ICs, I can use the others for RF projects too.
>slow to update from arduino
I planned on looking into the commonly-used library anyhow since it's just I2C, in case that's where the issue lies. Still, 1s is a long time. Since it's a sequencer I'll be updating the notes to a schedule, but I'll definitely want much less than 1s between each note. Do you know if it's a caching thing that won't allow 0.2s updates, or just a propagating delay that will? It has more than one output, so in a pinch I can hot-swap those with a muxer for 3 times the speed.

Unless you have a better suggestion than an Si5351 that's common enough to have good 3rd party documentation.

>> No.1948833 [DELETED] 

The human hearing is also limited in how it can detect tones. For most people 0.1-1% is what you can do. The capacity to distinguish BETWEEN tones is better tho

>> No.1948834

The human hearing is also limited in how it can detect tones. For most people 0.1-1% is what you can do. The capacity to distinguish BETWEEN tones is better tho. It seems you are trying to shoot a fly with a cannon. But if it works it works I guess

>> No.1948866

interesting read

>> No.1948943

heater fan. i bought some tools. got a heat sensor i can't find on the internet i think. got my clips and i'm bypass it and see if it runs. if it does i will order another type cut the wires and put different ends on. sound ok?

>> No.1948946

nothing happen but thats all it could be. look up more on it i guess.

>> No.1949247

yo I have one of those. I'm pretty satisfied with it personally, though maybe I just have low standards for oscilloscopes because most of the scopes I've ever had have been 40 years old from out a dumpster.
$150 is a really good price for it. It might not entirely outperform the $300 Rigol, and it can't do protocol decoding, but it's a pleasure to use and it does a lot. It was a good instrument in its time and still is today.
Probe sets for the logic analyzer can be expensive unless you get creative, so if it comes with those, great.

Be warned that the front panel buttons, especially the softkeys, get dirty and ineffective with age. The first thing you might have to do is take the front panel apart and clean the contact pads on the buttons.

>> No.1949270
File: 110 KB, 608x672, 2 d.png [View same] [iqdb] [saucenao] [google] [report]

Probably one of the dumbest questions.
What is the point of a D flip flop going into another? The text says that it is so that the output of flip fop r is a replica of the output q but delayed by 1 clock period. I can understand this bit but is there a reason why people would do this in a real-life scenario?

>> No.1949271

How do I connect a 10u cap to usb power if there's a sot23-5 tvs diode right next to the usb? after the diode? to the side on the vin side?

>> No.1949272

Shift registers.

>> No.1949274


I'm new to electronics and I want to learn. I also want to grab a deal in aliexpress, which is why I'm asking now:

- Any good starter kits (with arduino-like boards) that any of you know?

- Between these two multimeters, which one?

(your recommendation)

I'm from a third world country, so I would prefer cheap (for starters) and international shops, like ali.

>> No.1949275

> deal in aliexpress
The sellers on Ali just raise their prices before 11.11 and then drop them to claim you got some amazing deal.

>> No.1949276

What's the deal with 11:11? Not the person you replied to but everyone is talking about it for some reason.

>> No.1949278

"singles" day. Not sure if singles as in not in a relationship, or singles as in multiple ones (or both).

It's a thing in China.

>> No.1949280


It's like Chinese Black Friday.

>> No.1949289

I'm sure they do, of course I'm counting deals considering that

>> No.1949292

They're both decent for the price, that's why they're both highlighted.
If you want something actually good, look at Fluke 17b+.
Btw, some other shit to get: >>>/g/78636376

>> No.1949309

thanks, I completely missed that.

The prices are pretty cheap (overall), but the shipping is killing me. Can you recommend me a basic combination to get started?

one anon replied to that post and recommended these:

Do you agree, as a grab-all?

>> No.1949313

I'm the author of the large list. These kits are also alright. You could get a kit and a few (different) devboards, but the random non-devboard parts I did include in the large list are mostly stuff that the kits omit, and yet are useful (I'd argue most of them necessary).

>> No.1949314


>> No.1949315

And yeah, incidentally I have most of that already (and lotsa more shit I didn't include), and what I don't, I did order today,

>> No.1949317

alright, thanks!

I think I'll get the sensor kit + arduino pack from the other anon post, the multimeter and another devboard from yours. Trying to get all of them in one shop so I spend less in shipping

>> No.1949329

There'll be much fun to be had.
On my end, I'm playing with the iCESugar, using the open FPGA stack (yosys+nextpnr). Verilog is fun.

>> No.1949370


also get a waifuduino: >>1948774

>> No.1949390
File: 2.66 MB, 2976x2976, IMG_20201108_2136365.jpg [View same] [iqdb] [saucenao] [google] [report]

Newbie looking for some advice. I got this LCD game a while ago as a non-working on a fling, and recently I've been studying electronics a bit more and figured I would dig it out.

The first thing I did was remove a ton of corrosion from the battery contacts. That was easy, but then I ran into a new problem where the screen doesn't look right. Elements are lit up or not lit up correctly, etc.

I got to wondering if it wasn't a problem of the connections being weak as I've seen some examples of people fixing LCDs with missing elements by tightening the screws in their frame, for example. Well I opened this one up, and the mainboard looks fucked to me. I've never seen a board come out of a device and look like this before, so my assumption is someone back down the line tried to fuck around with it and fucked up. I'm not sure though. Can anybody give me some advice on how I might address this?

One thought I've had is with the LCD. The way this works is similar to another one I've taken apart; The display sits on top of two long rubbery pieces that contain circuitry, and the rubber is stuck onto the board where the contacts are for the display. I was wondering if I couldn't unseat this, clean off the contacts and the rubber, and reattach it. My concern is that I'll never be able to reattach it properly/it will be loose. The last device I did this on was smaller and screwed together tightly, so it still worked. I don't know that it would work on this one as it is more open and the LCD is not tightly held in place by the casing.

Any insight is greatly appreciated, thanks all.

>> No.1949392

>two long rubbery pieces that contain circuitry
Those are called Elastomeric connectors, or Zebra connectors.
>the LCD is not tightly held in place by the casing
this may be part of the problem.

>> No.1949395

Is it connected via a zebra strip? Try reseating it a couple of times to see if that helps, cleaning too. If they're misaligned or not pressing together with sufficient force, they could easily show erroneous images. Check the backlight too maybe.

>> No.1949398


looks like some of the PCB traces got exposed and someone use liquid paper to cover them up. this is completely unnecessary, of course. but it possibly hides the fact that the painted traces were broken. so you'll need to do continuity tests on those.

and yes, by all means, remove the LCD (tho i dont see how) and clean the zebra connectors with alcohol, and the PCB traces beneath them with an eraser.

it's possible the LCD is just being held in place by the zebra connectors acting like glue. i've see this before. presumably the screws at TRHS and BLHS are actually meant to hold the LCD against the PCB, and the alignment is done by plastic in the front panel.

>> No.1949403

That's the ticket. Thanks for putting a name to a picture. Further inspection suggests I was incorrect; There's some force that keeps the LCD on the board, but it's also sandwiched together, just in a bit of an awkward setup.

My fear is that the connectors seem to be fastened only by some adhesive force. I take them off and clean them, that I probably will struggle a lot to re-seat it all correctly. I wonder if there's a possibility that I could re-enable whatever adhesive force could have created this seal. I could definitely clean them up, just not sure it would reassemble cleanly.

Your assessment seems correct, as my second inspection shows. There's a plastic frame in which it all sits that the mainboard is fastened against inside of the unit.

The last time I took one of these off, I just had to carefully unseat the connectors, which went off without a hitch. I imagine this would be the same, except like last time, I assume it would be hard to get back into position.

I do have a multimeter so I should be able to test connections, I just don't know anything about how it should go together. You wouldn't happen to know a resource on learning to test boards that you don't know, would you?

>> No.1949421

I got some EPROMs a short while ago and don't yet have a way to program them. Now I know I can just use whatever MCU dev board I have lying around to make short work of it, but I'm thinking of doing it in discrete hardware just for a fun project. So using a CH340, I'll spit serial data to it via python or command line or whatever. I'll need no more than 3B/24b for this EPROM, as it's got 8 data pins and 13 address pins, with 3 more left over. I plan on shifting three bytes into three 74HC595 shift registers, and triggering a write just after latching them. For the timing, I'll use two 555 timers, one to trigger on the start-bit and turn off after the stop bit, and another at about 8 times the speed of the first to clock the registers. Then some way of checking when the data is all the way to the end. I could use a counter, but I think I'll reset the 595s after each write and use one of the last bits turning high to signify that the data has made its way to the third register.

Now this doesn't cover the verification of the data, but without them it's about 5 ICs, which is plenty small enough to be a little project.

Sounds fun, right?

>> No.1949424

If you are that bored.
Otherwise, look up "flashrom" and "MiniPRO TL866xx". Either will be able to help.

>> No.1949446
File: 302 KB, 393x600, launchxl-cc2650_launchxl-cc2650.jpg [View same] [iqdb] [saucenao] [google] [report]

I want to fuck with TI's CC2650 to make some BLE devices.
I bought pic related so I can start plugging peripherals to it. Eventually will design a circuit around the mcu once I kink out the other details.
The thing says it has an internal DC-DC converter as well as its own battery monitor.
Am I correct to suspect this thing is idiot proof in that I can just plug a LiPo battery to it and have at it with no external power electronics whatsoever?

>> No.1949459

Not an answer, but still somewhat useful:

>> No.1949548
File: 20 KB, 646x533, falstad.png [View same] [iqdb] [saucenao] [google] [report]

Can someone explain this particular situation? How can there bee current with no voltage?

>> No.1949552

all of the voltage is expended to create that current

voltage and current are essentially just two currencies for the economy known as "Power".
Power-as-in-Watts is the name of the game when it comes to voltages and currents, since P=V*I for linear circuits.

Your total voltage and current are determined by the power your supply can deliver. So, to generate a current, you must spend some of your voltage to get it.
Likewise, to create a voltage, you must necessarily decrease your current.
Maximum current happens when voltage equals zero, since P/V=I. When V=0 you sort-of-kind-of divide by zero but that's ok since your current is limited by your total power.

these simulators are misleading since they never take power into account and never show the real voltage sag that would be encountered IRL

>> No.1949571
File: 61 KB, 500x427, apu skateboard ramp.gif [View same] [iqdb] [saucenao] [google] [report]

Can I put a few Y rated safety capacitors in parallel? What are the drawbacks? I'm working on an old record player, and there is a capacitor that connects the neutral line to ground that is rated at .05uF, and the highest value I have is .01uF.

>> No.1949577

Use the 10nF cap you have and stop worrying about it.

>> No.1949608

add 1 ohm resistor between measurement points and behold the magic

>> No.1949646

Have Soviet 12V/24V NVG coming in the mail. Want to trash the buzzy transformer and replace it with my own. The tubes can take 20kV, and the transformer draws just 50mA or so. Is there an off the shelf solution I could use here? All of the plugs are Soviet proprietary type and have degrading insulation, so really I just need a rechargeable battery & transformer system with a fast blow fuse for safety or something. Then I’d wire it up to the goggles themselves.

t. Dumb mechE

>> No.1949650

Maybe a voltage multiplier?

>> No.1949660

Hi DIYAnons, i have access to a 40' led TV with a dead PSU: turns out samsung did shit tvs for a while and in the EU we don't have class actions and all of that stuff to have a new one for free and i checked online: those PSU boards really die on every aspect from caps (reasonable) down to transistors and voltage regulators (unacceptable and not worth it to start trace the board backwards to fix and replace every shit component samsung used with black friday deals close)
So i was actually wondering what i can do with it now, i saw you can just use it as a big led panel and i guess i could just directly wire up a PSU for the backlights right? or can i actually get more out of it by butchering the TV logic board and make it function as a big version of those HDMI screens for a raspberry pi?

For anyone wondering: even just downgrading it to led panel or hdmi screen would be great, the TV isn't actually mine and i'm actually discussing with the owner to just give him my older 50' fullHD TV for a good price to get myself a nice 4K screen (he doesn't really need a smart TV or 4K screen, he just watches it some times and the channels are barely fullHD anyway it was more for upcycling it instead of having him throw it in the dumpster

>> No.1949664

Get a working PSU, recap it, and change all the regulators and transistors with higher quality components.

>> No.1949710

you can look for the specs of the psu they have onboard and order a similar one ? What is the power rating and outputs?

>> No.1949713

i don't know man, it's 100+€ for one of those and will probably meet a similar failure after a while, i did inspect the board and caps are fine, the main relay just clicks on and off on it's own even when not connected so it's faulty logic on the IC parts and that i'm not even sure i can replace

>> No.1949731

I didn't know they were called zebra connectors, but if the display or the front panel buttons on your Fluke multimeter ever stop behaving weirdly/stop working altogether, this is likely why. I resurrected my Fluke 12 by cleaning these and reseating.

>> No.1949734

>have junk
>what do?



>> No.1949761

He is actually trying to fix/reuse something..

>> No.1949764

yeah, what about no? my father wanted a led panel for a studio light, if i can't have it work as a TV i know i can tear it down to the led and chassis and have a nice 40' led panel by just adding a powerjack and using a 12V brick, i can even create a tripod mount using the threaded inserts used for wall mounting it

>> No.1949784

What's the part number for the PSU?

>> No.1949785

Because that's an ideal wire, and ideal wires have 0 resistance. Since V = IR, as resistance drops to 0, so does voltage. In reality wires have small amounts of resistance, but simulators rarely deal with that.
Also superconductors exist.

It's just a PSU, doubtless you can cobble together a few PSUs from computers and such with the relevant voltage rails. Not like it's got a CCFL backlight, right?
Is the existing control board seperate from the PSU board? Does it have issues too?

>> No.1949815

i found a guy that had a similar problem+a row of the backlight didn't work, he had to replace all caps (easy) but that didn't fix the problem so he actually had to replace one of the ICs that regulate power to the leds, i wonder if it's something similar (my backlight works) and that same IC somehow has to turn on and off the main power stage for the TV to work....how cheap can i go for an hot air soldering kit? i'd use it only once but i don't think it's possible to do such a work with a cheapo iron like mine (or is it?) also i guess i should first search if i can find those ICs alone...i'll try keep you anons posted if it's intrested but maybe it can actually be fixed

>> No.1949822

Look here first.

>> No.1949829

Here's a kit with the components for the repair for £25.

>> No.1949836 [DELETED] 

Update on the AC circle saw:

-I got continuity over the brushes and the motor budges a little bit every time you give it DC power. It stays absolutely still when plugged in and the switch is pulled

-On the power cord, I get continuity from one of the plug's blades to the motor brushes (when the switch is pulled in), but not from the brushes to the other blade. Is that a sign of a bad switch or internal short, or is the other blade supposed to power the stator magnets? (I can't remove the wires from the switch to test the switch without cutting them. the switch has two "line" wires and two "load" wires)

>> No.1949839
File: 32 KB, 456x730, ninty.jpg [View same] [iqdb] [saucenao] [google] [report]

Update on the AC circle saw:

-I got continuity over the brushes and the motor budges a little bit every time you give it DC power. It stays absolutely still when plugged in and the switch is pulled

-On the power cord, I get continuity from one of the plug's blades to the motor brushes (when the switch is pulled in), but not from the brushes to the other blade. Is that a sign of a bad switch or internal short, or is the other blade supposed to power the stator magnets? (I can't remove the wires from the switch to test the switch without cutting them. the switch has two "line" wires and two "load" wires)

Pic related is the switch

>> No.1949855

>I get continuity from one of the plug's blades to the motor brushes
The switch only needs to switch one side of the circuit, that's normal. Is that a dimmer switch, or just an on/off switch?

Don't be afraid of cutting the wires, provided you've got a soldering iron and some heatshrink that is. It's going to be difficult to test anything without isolating the parts. If you have to replace something you'll need to cut wires anyhow.

>> No.1949860

>a dimmer switch, or just an on/off switch?
I forgot to test it for variable resistance but it has a smooth, continuous pull so I think it's the former.
>Don't be afraid of cutting the wires, provided you've got a soldering iron and some heatshrink that is.
Would electrical tape work?

>> No.1949861

>I can't remove the wires from the switch to test the switch without cutting them

you can test continuity by sticking two needles or pins through the wire insulation to get at the copper inside. when you remove the pin, the plastic sleeve should ''heal'' and the hole disappears.

anyway, it definitely sounds like the switch is burnt. possibly only 1 of the 2 poles are burnt, so you can short that pole, and still use the other pole to switch motor on/off.

>> No.1949863

>variable resistance
It will be a phase-fired-dimmer, using a TRIAC as a switching element. No variable resistance will be measurable from the outside.

Replacement switches are easy to find, but considering it budges for a quick second but does nothing after, I'm thinking it might be a seized motor or something. Does it rotate by hand?

>> No.1949869

Yeah, the motor rotates by hand. I was giving it power from a 30v5a lab bench when it's rated for 120vac10a so it could be that I'm underpowering it. It keeps rotating (slowly) if I keep tapping the PSU leads on the contacts.

>> No.1949876

>I was giving it power from a 30v5a lab bench
What, directly to the motor or via the switch? Because the switch won't do shit on DC if that's what you're doing.

>> No.1949878

Directly through the brushes.

>> No.1949882

>Directly through the brushes
That should be working, but remember you're backfeeding power into the TRIAC driver, which might be interfering with your measurements. Probably not, but if it gets hot or your PSU starts current limiting, you know what's happening.

Considering the motor turns by hand but not by electricity, I'm a bit stumped as to what the issue could be. If you hook up the DC power and spin the rotor by hand, does that kickstart it? Could be that the starting torque is insufficient to get it going at that voltage. Or just try it again on mains.

Else, check the wiring of the comm-bars to ensure nothing's broken off, like I mentioned earlier.

>> No.1949884

>your PSU starts current limiting,
IIRC it started current limiting and the voltage started dropping.

>> No.1949889

Then either it was shorting through the TRIAC speed controller (unlikely now that I look at a circuit), or the resistance of the windings was too low. Either way it isn't a particularly good test of the motor.

>> No.1949898
File: 1.78 MB, 3264x2448, IMG_9463.jpg [View same] [iqdb] [saucenao] [google] [report]

I have these shitty old koss computer speakers with a blown amp/pcb, can i rewire it so I can use both speakers together, (unpowered) or just toss it in the trash

>> No.1949899

They'll be pretty quiet without an amp, but they will work. Better yet, just get a cheap amp board from aliexpress or ebay or wherever.

>> No.1949977
File: 924 KB, 1312x476, dualheadlightmod.png [View same] [iqdb] [saucenao] [google] [report]

Hey guys needing some guidance, I think I have an idea on what to do, but would like other's opinions. I have a E-scooter, a nanrobot D4+ 2.0, and it has this lousy single headlight near the bottom. (left scooter is the model I have) and Newer models of the nanrobot some have dual headlight like the right scooter (nanrobot ls7)

I am wanting to remove that single light, and go with a dual headlight setup like LS7.

So, I don't want to burn my scooter up when doing this mod, and I would like to know if checking current power draw is all I really need to know if this mod will be safe to do.

1. Check voltage/wattage of the current headlight installed.
2. check the milli amp/amp draw (is this possible on a digital volt meter?)

are these really the only two things I need to check to make sure I have enough power/current?

These are the dual headlights I am wanting to buy.

the power is 15 watts (per headlight I am assuming, so 30 watts)

Would going series/parallel depend on the power ratings? One guy did it on his D4+ and I asked him, but didn't speak english well, and asked if he did parallel or series, but never got a response.

Should I consider using a fuse on each headlight?

Hopefully my post makes sense enough to help me out, Thanks!

>> No.1950009


dont do it. 1 light is better than 2 lights. say you have a squirrel running across the road. with two lights, it casts 2 diff shadows and confuses your brain so much that are left paralyzed to do what's necessary: brake and curse, ''Fucking Trump, this is all your fault.''

plus, you'll never make it look like it's stock. it'll look nigger-rigged, and the grade 6 girls wont be asking for a ride.

>> No.1950022

R = V/I
R = 0 / 19mA
R = 0
You got a nice conductor there.

>> No.1950023

I didn't ask if I can make it look stock.

Also let the squirrel run across the road, I'll hit it regardless.

two headlights is preferred that I can use strobe functionality to catch people's attention when riding.

I'll just go wing it.

>> No.1950029

yeah you can use a multimeter to measure the volts and amps, just read/watch a guide if you don't know how to rather than guessing.
Definitely do parallel, only very rare cases are these kinda things done series. You want consistent volts on both even if their resistance wavers a bit.
sprinkle fuses everywhere so you don't pop your power supply when you short to the chassis.
poor english guy probs didn't want you to blame him when you broke your thing cause you sound like you're over your head, but good luck famalam, hacking on your stuff makes it feel so much more like your stuff.

>> No.1950033

In over my head? Eh maybe, but as long as I am being cautious on what I do, it should be fine, well in a perfect world "it should be fine"

my "knowledge" on electronics is bits and pieces, scattered in different areas, not consistent knowledge.

>> No.1950034

If your dubs are any indication you've got the luck you'll need.

>> No.1950042

I got dubs squared. Shit famalam

>> No.1950068
File: 159 KB, 1280x720, hosico.jpg [View same] [iqdb] [saucenao] [google] [report]

Lurked here for almost one year, now got a job interview appointment for junior hardware engineer (PCB designer).
Please give me power, most company still don't care about KiCad and I know nothing about Altium.

>> No.1950073

any one know what happens if i charge a capacitor with a resistor in series with it from a boost converter and then drawn from the capacitor alone .

will i get the voltage it charged at as if from source like a battery or will it still be resistance dependent ? i want to say the voltage is stored and the terminated cap will discharge the stored value not counting any spike from initial connection

i have seen things demonstrated but i dont know how accurate they are and i know theres some bullshit with where the cap goes and if i should have series or parallel resistance on circuits. but my question is simply will the cap store the voltage value when it was charged with series resistance and discharged without it

>> No.1950078

>but my question is simply will the cap store the voltage value when it was charged with series resistance and discharged without it

yes, as long as you dont wait too long coz caps will self-discharge completely in hours or days.

>> No.1950081

>Please give me power

give yourself power. show up with color photos, or actual PCBs, with some of your best designs. tools can be learned, but talent is inherent, and easy to see.
or maybe you're a shit designer, that's easy to see as well.

>> No.1950088

Need a benchtop power supply
Two+ outputs, and decent build quality, nothing too fancy
Any recommendations, lads?

>> No.1950093

>give me power
P = I*V = I^2*R = V^2/R

>> No.1950106

I have this one for my bench.


>> No.1950110

Was looking at this one
Though I suppose I could just go for two cheaper single output supplies

>> No.1950140

Why is this /diy/ instead of /g/?

>> No.1950141

/g/ is about technology products, we build shit or "DIY" shit

>> No.1950142

I'd argue that programming is as much diy as embedded development, but that general is still /g/

>> No.1950150

my opinion doesnt mean im right.

>> No.1950151

/g/ is not only programming. It is mostly consooooooomerism, part swaping and other stuff.

>> No.1950155

It's unfortunate, but /g/ is mostly a plebsumer board. This is a maker board.

>> No.1950169

I know you mean well but fuck do I hate that word and the culture surrounding it

>> No.1950170

If you prefer hacker, that's welcome too.

>> No.1950189

Why are BLDC motors so unpopular compared to AC motors with VFDs?
VFDs need DC conversion anyway so does it really save anything?

>> No.1950244

induction motors are cheap to make and repair. You can also buy the motor and install a VFD later to improve efficiency. I guess when magnets become cheaper and power electronics even smaller the situation will change. I am talking about industrial use here (compressor,pumps etc) no idea about what area you are talking about.

>> No.1950245

> You can also buy the motor and install a VFD later to improve efficiency.
in the sense that whatever you are powering can run ok at 100% power and fixed speed or has other means to control (variable displacement displacement in the case of pumps ,unloading valves for compressors or gears for other shit)

>> No.1950253

I am indeed talking industrial use but areas where speed control is important.

You see BLDCs in so many things that run off batteries like cordless tools, electric bikes and cars. So I'm wondering why they're rarely used in situations where AC mains power is available since the VFD converts to DC before inverting it again anyway. Or inversely why do battery powered machines not use VFDs with AC motors?

Is it just because energy efficiency trumps other concern there? Are there problems with higher voltage BLDC motors?

>> No.1950259

Well I think it is mostly because of power rating.
I do not think it is easy or cheap to make a permanent magnet for a +2 MW motor. On the other hand the rotor for a induction motor will just be a lug of iron and aluminium.
>Or inversely why do battery powered machines not use VFDs with AC motors?
Why do you think they should?

>> No.1950261

how do you draw 1a from a battery?
example a aa or 9v battery how are you able to tell it how much current to use?
doesnt it operate at a current?
i understand that you could use resistors to limit the current but how do you make it supply more current?

>> No.1950262

what switch is it called when it is a momentary switch but it does not stay closed when held in the same position because the switch opens itself after being closed

>> No.1950263

how can you convert ampere turns to magnetic attraction force in joules or kilograms or kilogram meters

>> No.1950264

draw the equivalent circuit of the battery and connect it to a load, calculate the Voltage and current at the load.
go home pajeet, no one will do your homework for you

>> No.1950267

Looking more into why electric cars use BLDCs (which I'm wrong about, some do use induction motors) I found an interesting discussion about it from Tesla:
Apart from the higher costs of permanent magnets it mentions concerns like this one:
>In contrast, induction machines have no magnets and B fields are “adjustable,” since B is proportionate to V/f (voltage to frequency). This means that at light loads the inverter can reduce voltage such that magnetic losses are reduced and efficiency is maximized.
>With DC brushless, as machine size grows, the magnetic losses increase proportionately and part load efficiency drops. With induction, as machine size grows, losses do not necessarily grow. Thus, induction drives may be the favored approach where high-performance is desired; peak efficiency will be a little less than with DC brushless, but average efficiency may actually be better.

>> No.1950269

I don't know what I can say...
I am dressed as a clown and telling a theatre that the building is on fire.
And they think it's a joke...
I (or you), need to speak to engineers.
Ones like me.
Do you think I am here to disrupt your shit? I might draw the schematics,
If you don't have what you should have in USA...there's something else.
Look at the push/pull solenoid.
Watch videos. Watch it break glass.
Let your mind wander.

>> No.1950270

cars are in the power rating where BLDCs are used. I was talking about +MW range. But as an exercise you could finding the price of a BLDC motor and a induction motor for the at the same power rating.

>> No.1950271

And they have DC already available too.

>> No.1950272

For MW stuff, I'm not sure about motors but there are brushless generators, which wouldn't be that different material cost wise to a motor, used in wind and steam turbines.

>> No.1950276

I know, I visited a Wind power farm once where they had a 1.5 MW synchronous generator. The added cost is supplanted by the alternative:
Shitty induction generators.
And synchronous machines are not BLDCs but you reach a point where the definition kinda loses meaning. SM are still AC machines, no DC link at all.

>> No.1950277

no do it

>> No.1950279

pls explain im stupid
does the battery have a current that it provides the circuit

>> No.1950308

then everyone will think you're just like in the movies, and thats cringey.

>> No.1950346

A battery has a voltage, the current is determined by the resistance of the load

>> No.1950390
File: 1.74 MB, 1939x2312, 2343673534563.jpg [View same] [iqdb] [saucenao] [google] [report]

Need help diagnosing AC motor, 120v 3hp
>starts and runs freely forever
>under load runs for a 60sec and trips thermal protection
>reduced load, (rebuilding air compressor piston, lubed, aligned etc)
>still trips after 1min
So the caps might be bad, 20µF 370v ac 60hz. Both the start and run caps show no signs of stress or bulging. Both caps charge and discharge within the limitations of my multimeter (resistance). I don't have a method of testing spec'ed µF capacity just resistance. How would one test them to determine if they are actually bad before troubleshooting other parts of the compressor?

>> No.1950411

>the current is determined by the resistance of the load
so with hypothetically zero resistance what is the current of the average AA battery providing (or the maximum that it could provide without needing to use another) in regular amps

>> No.1950419

latching switch

Attraction force measured in newtons, or torque measured in newton-meters, can be calculated relatively easily if you know the magnetic field strength and the magnetic dipole of the thing being attracted. A magnet with twice the dipole of another will experience twice the force. It's also a vector operation, so it depends on orientation. While it's probably quite difficult to calculate the dipole strength of a permanent magnet, you can make an educated guess based off its size and surface B field, which should be sufficient. The calculation of the magnetic field strength from the solenoid is a trivial task and has been left to the reader as practice. It's all on wikipedia and/or hyperphysics.

If it starts fine then the caps won't be too low on capacitance, I'm thinking it's something else. Is there a fan on/in the motor? Is its airflow impeded by dust? Where is the thermal switch/fuse located?

With hypothetically zero resistance, a battery could provide infinite current.
However, in reality, a battery can be approximated as an ideal voltage source (can output infinite amps) in series with a resistance. See Thevenin's theorem. The more current you try to draw, the more the voltage droops. If you try to draw I = V(no load)/R(series), then the output voltage will be 0V.
I'd look up a battery datasheet to see how much ESR (equivalent series resistance) they actually have, or their maximum current value which may be lower than what you calculate from ESR due to thermal reasons. It will vary depending on whether you're using alkaline, dry cell, lithium, etc. chemistries, so go for a datasheet for the specific batteries you have.

>> No.1950422

>latching switch
latching switches are on/off normal switches arent they? like light switches, or those clicky buttons like on pens.

>> No.1950423

>so go for a datasheet for the specific batteries you have.
ok thanks

>> No.1950426

can you run a very efficient electromagnet on near 0 volts

>> No.1950433
File: 69 KB, 1060x455, duracell ultra duralock power preserve ultra power.jpg [View same] [iqdb] [saucenao] [google] [report]

is this for fucking real there is no way the aa battery could do this

>> No.1950435
File: 43 KB, 804x381, duracell ultra duralock power preserve ultra power.jpg [View same] [iqdb] [saucenao] [google] [report]

i am clearly still learning to read

>> No.1950443

the opposite of a "latching switch" is a "momentary switch"

See: MRI machines

either way it would overheat very quickly
you want to look for an "absolute maximum" value of some kind

>> No.1950455

Resistance can't be any lower than the internal resistance of the battery. Note that "industrial" NiCd batteries can realistically source >100A if short-circuited, which is why they aren't sold to the general public.

>> No.1950457

yeah I've seen nmr coils run at 0.2V and 500A, and that was only ~4 tesla

>> No.1950471

retard posting hours?

>> No.1950474
File: 26 KB, 1482x514, dPV.png [View same] [iqdb] [saucenao] [google] [report]

I am planning to build a board that simulates PV panels, ideally it will have three "PV circuits" like pic related in series I can show the effects of shading and bypass diodes. My question is, what kind of circuit can I use to implement a current source like that?

It is for an educational thing for technical students/hobbists/tradies

>> No.1950475

Even a puny consumer AA NiCd cell can do like 20A.
Put some in parallel and you can really light a fire.

>> No.1950484
File: 78 KB, 720x448, pov.jpg [View same] [iqdb] [saucenao] [google] [report]

Welp, never thought about using a buck/boost converter by controlling the current instead of the voltage. I guess that is it folks. If you have any other suggestions I am open

>> No.1950494
File: 48 KB, 1615x582, paraPV.png [View same] [iqdb] [saucenao] [google] [report]

this fellas canno tbe put in series, only in parallel so I guess I will pick a buck IC and make some simulations. /endblog

>> No.1950523
File: 80 KB, 1318x701, img_FgZw6mVBo8WgigH_r.jpg [View same] [iqdb] [saucenao] [google] [report]

I have no idea where else to go with this question. I have two giant sansui 3-way speakers. I moved into my gfs house recently and found an old stereo receiver, technics 'SA-GX710'. I hooked them up and they sounded really good for a couple weeks. Then something strange started happening. Without changing any of the channels, tuning or cables, ect. Basically the issue started slowly but then got progressively worse over time, a dull and quite 'fuzzy-buzzy' sound started coming through. It started soft but now it's really strong and completely ruins all enjoyment. It almost sounds like 'clipping' but it happens continuously and it's not as much of a sharp snapping noise, which usually comes from clipping but rather a more dull and fuzzy sounding buzz. I have no idea how to start debugging this issue, can anyone point me into the right direction. One interesting thing I noticed is when I unplug one of the white/red aux leads, the issue gets considerably better, the buzzing noise almost goes away completely, but I'm not going to just settle for one speaker. The issue also comes and goes, sometimes it completely goes away without messing with anything during the duration of the same audio track, it's very strange. Can someone please help me. Sorry, I posted this in the wrong general earlier.

>> No.1950546


old stereos, like old people, get sick, then sicker, then die. go to thrift store with $10-$15 and get another one. dont waste precious time replacing caps and transistors. those can delay death but not prevent it.

>> No.1950561

Completely and utterly retarded. Quality old stereos can live for 40y in a damp shed with no servicing and be ready for another 20 years bare minimum. Death cannot be prevented for anything in this damn world but going from dead to still playing 20y after is not what I would call a mere delay.
Here your problem is not whether or not it can be fixed it's if its worth it at all. This kind of old shitty AV amp is not worth a penny nowadays and can be swapped for literally anything else you find at the thrift store. If this were a vintage 70s reciever I would have considered a repair but here its not worth it.
The problem you describe could be coming from literally anywhere and you let it sit long enough to get from bad to worse. I hope you wont do this with something actually valueable. As always with this kind of equipent n°1 suspect would be dead caps. Bad filter caps give shit power and if shit comes in shit comes out. You could have a leaky (as in leaks DC) cap somewhere that slowly but surely killed a transistor or even a bias pot that's intermittent. Not worth chasing the issue. Probably a dead transistor imo. Scuttle the amp for parts, throw out the caps and move on.

>> No.1950609

okay thanks, I'm getting another (old) receiver out of storage at my dads place sometime, I'll swap it out and see if it fixes the problem.

>> No.1950611

Also it possibly might be a ground loop.

>> No.1950649


>> No.1950675

No thats retarded. Ground loops dont cause distortion and they dont get worse and worse as described. They are also not intermittent. Im 80% sure its a transistor.

>> No.1950679

It does indeed. One speaker vs two is what's interesting. As the channels are always kept separate throughout, it has to be power-related.

>> No.1950737
File: 115 KB, 427x960, 1601665116461.jpg [View same] [iqdb] [saucenao] [google] [report]

For the ground loop issue, from just googling 'ground loop', and reading a couple random articles I'm going to guess that's definitely not the issue. The speakers don't make a continuous buzz noise like is described in the articles, it's only during certain portions of the audio. And I think the single speaker channel thing is a nuance which could be the biggest clue to understanding what's wrong. Oh well I'll probably just get a whole new stereo, I don't even have the skills or equipment to fuck with something on that level.

>> No.1950740

You could always get headphones.
Sennheiser HD600 will beat most speaker setups out there.

>> No.1950744

Yeah, I already have some nice iems and sennheisers. I want my speakers too, I like blasting music loud enough that it forces my gf and neighbors to listen to the same song I'm listening to, because I'm a psychopath.

>> No.1950792

I'm planning on building a SMPS to boost a low voltage source to >200V at a fairly high current (for the voltage) of around 100mA. Due to this I need some pretty high peak primary currents of >10A. My issue is my input supply is only 3A capable.

What I'm wondering is, do I REALLY need a supply that can match the peak primary current I need, or, will the input storage capacitors provide that current? I'm switching at about 100kHz and calculated a switch on time of slightly under 5us. Here's the thing though, I found the A*s product during the switch on time to only be something like 125pA*s where as the total A*s product for a single fully charged 220uF input was like 2.6mA*s. Based on my calculations supplying 10A for like 4-5us should only drop the capacitor by a few mV. Basically fuck all. The cap charges back up in the switch off time and bam, I can get 10A peak current from a 3A supply.

Is this analysis correct or am I not understanding something?

>> No.1950832
File: 21 KB, 425x360, 41nhoRy-1iL._AC_SX425_.jpg [View same] [iqdb] [saucenao] [google] [report]

Anyone have experience with those cheap dc dc XL6009 step up converters? I need to power an old netbook at 19v at 2ampers. Already have a 15v 5a lab psu.

Some reviews say these converters shit the bed as soon as they go over 10 watts.

>> No.1950861


95% sure your 15V supply will work just fine. i've powered plenty of 19V laptops using 16V IBM power bricks (which cost $2-$3 at the thrift store) without a hitch.

>> No.1950889
File: 51 KB, 1185x769, untitled.png [View same] [iqdb] [saucenao] [google] [report]

Guys, please tell me this will work

>> No.1950890

What do you use for circuit and PCB design lads? I use kicad but recently tried EasyEDA on insistance of a coworker. It's pretty good and the part marketplace integration is to die for. Kicad PCB editor is still way ahead though

>> No.1950897

no one is googling what the fuck those letters mean, draw a proper circuit diagram or explain yourself a little bit better.

>> No.1950898


>> No.1950901

I want to make a solar panel charge the battery and also supply the load, but I want to avoid using MPPT controllers as the they are too expensive or PWM controllers as they are too inefficient, LTC3780 is a buck/boost converter and MBRF20100CT is a diode.

>> No.1950902

Good luck then in drawing your circuit diagram

>> No.1950903

It will work :)

>> No.1950905

are you sure?

>> No.1950906


>> No.1950918

PM'd you

>> No.1950919

Thanks, I will check this out. May help with the programming side (I remember jack shit since I graduated)

>> No.1950924

use the scope and not probe. The probe rounds off small values.

>> No.1951029

Just get the PWM controller and panels of the appropriate voltage, the efficiency is just fine under normal lighting.
MPPT is better because it compensates for different lighting conditions to keep the panel at max efficiency, but your schematic doesn't do that so it will probably be no better than PWM, and possibly worse.

>> No.1951034

Alright, thanks man I'll resort to PWM controller then. MPPT controllers are very expensive what the actual fuck

>> No.1951044

You use MPPT when you are making a fixed installation. Inverters/converters for PV already have some sort of MPPT in them.

>> No.1951050

Not related to fact that they are fixed, it is more about power x cost. But if you are interested is not hard to make a perturb and observe MPPT thing with arduino.

>> No.1951057

Unlike PWM, MPPT controllers need a big fat inductor, which means lots of copper and costs many monies.
I'm sure they charge a hefty premium for the 20 line microcontroller program also, because they can.

>> No.1951058

Sorry I am talking about MPPT as an algorithm. I do not know what you mean by PWM controller. You could implement MPPT in a regular buck/boost converter. The inductor is sized according the frequency and power rating of the converter.

>> No.1951063
File: 20 KB, 600x600, Untitled-7-1.jpg [View same] [iqdb] [saucenao] [google] [report]

>I do not know what you mean by PWM controller
It's what they call those cheap solar charge controllers that really just disconnect the panel when the battery is charged or it's dark. They're dirt cheap because it's just a mosfet and a microcontroller that decides when to turn it off.
>You could implement MPPT in a regular buck/boost converter.
Yes, but to make it worthwhile your buck converter needs to be very high efficiency over a wide range. And that means lower frequency (to reduce switching losses) so high inductance, and also low resistance inductor.

>> No.1951102

Guys, I think BQ24650 is the most cost effective solution to my problem. It is a pretty efficient MPPT controller and I only need less than 5A anyway. What do you think?

>> No.1951131

So as you showed yourself, the controller can have MPPT regardless if is using PWM,PFM or your mom. (Those are modulating schemes, not control ones.) Good luck anon, post results later this thread needs more stuff other than indians asking for homework solutions

>> No.1951210

yes, it ought to work fine. The cheap modules on ebay have no screens, but you could just use a multimeter to check how it's working.
Technically yes, but the solar-marketing-bs industry has adopted PWM to basically mean "not MPPT". I think the PWM they are actually referring to is the trickle charge mode many controllers will go into when the battery is full, but even if it doesn't have that feature people will call it a PWM controller.

>> No.1951253

how much did you pay for it?

>> No.1951277

Today I failed to desolder a capacitor. It was the only thing stopping an old switch-mode PSU from working. I tried dry, with extra solder, with and without flux, yet the original joint just would not melt. It was an extremely humiliating experience. Was it the ground plane? Was it some stubborn SN100C alloy? In any case, my trusty, old, plug-in-wall Ersa was outdone by the chinks. I am buying an actual station.

>> No.1951302

What's a good online electronics store for northern Yurop? Not for components, but tools and supplies like flux, solder tips, etc. I can't stand the long delivery times and risk of fakes from chink stores.

>> No.1951305

hobbyist noob here.

I can't be bothered to wait another two months to have a JTAG board sent from china, so I want to use an arduino.

Here is the library I plan to use. https://github.com/mrjimenez/JTAG

He uses voltage dividers to take the 5v of the arduino down to 3.3v for the jtag. Does this sound like a good idea? If so, I have a voltage divider set up for 2.7v. Will that be OK? I apologize for asking basic questions. I'm paranoid about ruining my fpga (I care more about the wait time from china for a new one than the cost).

>> No.1951306

Pretty good lectures on digital logic

>> No.1951310

I think for Europe in general, Farnell is one of the more local ones, as opposed to American owned ones like Digikey. They do both tools and components IIRC.

>> No.1951314

There's also RS components.

>> No.1951321

For 16.6, $0 courtesy of the pirate bay.
I also have access to the latest version through work, I can just remote in and use that if I wanted but I generally don't for hobby projects. No idea how much they pay for the licenses.

>> No.1951326

Dont use Farnell for anything above 50 euro. These niggers are way more expensive across the board and their selection is godawful. When I had a big project I tried to make the same cart over at mouser and not only did I have more options It also nearly halved my bill. As soon as you get above 50 euro mouser writes off the shipping and at this point farnell just isnt worth it. Under this the shipping costs will make Farnell the cheaper option.
Also dont buy multicomp unless you dont give a shit about quality.

>> No.1951329

Just crank it up 50C noob

>> No.1951335

>As soon as you get above 50 euro mouser writes off the shipping and at this point farnell just isnt worth it.
"Free shipping" means you still need to pay duty, VAT, an extra handling charge for ther import that DHL, FedEx, etc. charge you (around 10€).

>> No.1951344

Oh, they don't have a free shipping thing? I've only ever used alibay because I never buy more than $50 of components at a time anyhow. IIRC Arrow has an even lower free-shipping threshold.

>> No.1951346

I obviously include VAT in my equation. Im no burger.
Just logged into my account to check.
Incoterms : DDP (droits et taxes de douane payés par Mouser).
When buying in france they dont make you pay imports unless you buy something that requires additional charges (you dont). I dont know how it is for UK or the rest of europe. Also Farnell's stock is decimated they are out of everything and they make you pay 18 euro per item number to import the shit you need or you have to wait till the end of december.
No they dont. Farnell make you pay additional charges under 30 euro and mouser under 50.
I hope for your sake you dont use anything you bought over there for anthing you care or important.

>> No.1951380

I looked at Farnell and maybe it's just my incompetence, but when I searched for 'flux pen', everything that came up was $100+, so I guess I'm not in their target group of customers.

>> No.1951387
File: 1.81 MB, 2340x4160, IMG_20201029_110408823.jpg [View same] [iqdb] [saucenao] [google] [report]

So i help disabled people and want to let them use the joystick of their electronic wheelchair to play games. I want it to be easy of use so they should just unplug the cable to te wheelchair controller and put it in de game controller. The big cable, pic related, has 2 power pins and 2 data pins. The joystick is completely programmable (like switching axis and lowering sensitivity). I can't find a datasheet, only a technic manual which doesn't contain anything of use. I tried just reading the voltage and resistance on the data pins but that gets me nowhere. Am i fucked?

>> No.1951396

>2 power pins and 2 data pins
Is it a USB interface?

>> No.1951399

Do you have a scope? Could be i2c or serial. Or it could be something completely different.

>> No.1951407

>I hope for your sake you dont use anything you bought over there for anthing you care or important
Depends on what it is. My resistors have steel leads but I don't care since I'm not making anything to last or go next to the ocean. My ICs seem to work just fine, though I mainly buy old shit like LM324s, not obvious knockoff STM32s. LEDs, LDRs, rotary encoders, switches, potentiometers, they're all just fine. I don't make anything to sell to anybody, so not having guarantees on my parts isn't an issue.

Get a cheap logic analyser:

>> No.1951425

I don't think so. It's an R-net system. I think they have a custom communication protocol

No scope sadly

Yeah i was afraid i have to get a logic analyser. I have no experience in this. Is the logic analyser good if you don't know the communication protocol?

>> No.1951432

>Is the logic analyser good if you don't know the communication protocol?
Logic analyzers are good if you want to see and understand what's on the wires.
It helps you if you know the protocol, as you can decode it outright; PulseView from sigrok will even do that for you, for a lot of protocols.
But if you don't know the protocol, it will help you figure it out.
Never a bad idea to get it as part of your toolbox of useful electronics. >>>/g/78636376

>> No.1951433

>The big cable, pic related, has 2 power pins and 2 data pins.
Likely some data packets on those 2 data pins, encapsulated in a common protocol like i2c.
The aforementioned nanodla, with its 24MHz sampling, will have no problem whatsoever with i2c or any other of those low-speed protocols.

>> No.1951437
File: 369 KB, 1080x1250, 1539286348094.jpg [View same] [iqdb] [saucenao] [google] [report]

All right i'll look into it. Thanks guys.

>> No.1951477

Yes as the other guy said, the free software PulseView has a bunch of protocols inside it that are relatively easy to test, but they do need some prior knowledge. Also those cheap logic analysers go only up to 24MHz, which is likely insufficient if that thing is USB, but still better than any cheap scope.
The fact that they've used a non-USB connector makes me think it isn't USB, as USB is about the only standard out there with standardised connectors for varying form-factors. On the other hand, I've seen people using PS/2 for RS232 because a D15 was too large. So just knowing what connector it is doesn't mean you know what protocol it is.

Speaking of RS232, if you want to try and get data from it in PulseView, you need to know (or guess) the baudrate. Because it's asynchronous; of all the damn pins on those D connectors they never put a clock, even just a discretional one. Pick whatever standard baudrate is closest to 1/T, where T is the length in seconds of a 1 or 0. RS232 is pretty easy to see by eye though, which helps.

>your toolbox of useful electronics
That's a fucking warehouse mate.
also i'm regretting not buying an easy cheapy badusb for usb fuckery on 11.11

Also I wonder if anyone's made a faster logic analyser with fewer channels just by connecting some coax for a delay between channels. Say you only want 2 channels instead of all 10, so you loop 3m of coax from 1 to 2, then 3m from 2 to 3, then 3 to 4, 4 to 5. Or use logic gate propagation delay I guess.
Could get 5 times the effective sampling speed like that.

>> No.1951479

>That's a fucking warehouse mate.
I actually ordered everything there. The parts are cheap, so the whole is not that expensive.
>I need more megasamples/s
You could get something else like the dslogic. That also has more channels, for that matter.
But the nanodla is less than $10, and extremely convenient vs not having an analyzer.
>Say you only want 2 channels instead of all 10
Some analyzers are actually bottlenecked by e.g. usb2, and have modes that allow higher sample rates with less channels.
Also, the less cheap analyzers typically have hardware triggering feature, and can capture into an internal buffer at very high speed, sending the capture to the computer later.
Buffered mode vs streaming mode.

>> No.1951482

>Speaking of RS232, if you want to try and get data from it in PulseView, you need to know (or guess) the baudrate
You don't need to "guess" anything. Sample at max sample speed. You'll see clear runs of n-bit sized low logic levels. Pick one of the shorter runs, measure the width, there's your bit size or period.
Invert that, you have your frequency or bitrate. Put a uart decoder on that signal, with that bitrate, you're golden.

>> No.1951484

It's 480MB/s, that's enough for a lot of stuff at only 1 or 2 channels, though of course not enough for USB. I'm now wondering if there are FPGAs or MCUs with native USB 3 or thunderbolt or whatever. A DIY PCIE card might be easier, though that's got its own issues.

I'm not talking about sample-rate, that you should turn up as high as necessary to get a clear picture without aliasing. I'm talking about PulseView's decoding baudrate.
As I said:
>Pick whatever standard baudrate is closest to 1/T, where T is the length in seconds of a 1 or 0

>> No.1951485

>As I said
Apologies Anon. Somehow I managed to read your whole post, except the second half of that long line.
>of all the damn pins on those D connectors they never put a clock, even just a discretional one.
Being async is a feature of the uart protocol, not a bug.
There's a lot of synchronous protocols to pick and choose from, if that's what you want. SPI and i2c are quite popular.

>> No.1951492

>Being async is a feature of the uart protocol, not a bug
I know and there's nothing wrong with that, but just putting a clock line there in case a device wants to use it for whatever reason would be handy. Not like anyone ever uses the damn DSR and DCD and RI pins anyhow. Why even use a 25 pin connector in the first place?
Actually I'd quite like to see what the earliest RS232 receiving/buffering ICs were doing, to push each byte onto a bus or store it in a register or whatever. They'd need their own (adjustable?) timing system too.

>> No.1951496
File: 1.70 MB, 1920x1080, 1598953706913.png [View same] [iqdb] [saucenao] [google] [report]

Ironically, I recently discovered the reason DOS serial port calls (freedos or msdos, doesn't matter) didn't fucking work no matter what computer or dos version I tried was DTR/DSR.
DOS serial functions do call bios behind the scenes, and bios expects rts/cts + dtr/dsr,
Ridiculous, but that's how it is.
Incidentally, Linux's completely fucked when it comes to DTR/DSR. You're literally better off bitbanging than trying to use Linux serial support.

>> No.1951498

Is that just for hardware serial ports, or do you get the same issues with a USB-to-UART converter or whatever?

>> No.1951507

usb to rs232 or motherboard backplate, it does not matter. Same shit. No DTR/DSR on Linux.
The way is to breadboard-bullshit those signals from RTS/CTS.

>> No.1951516

>Not like anyone ever uses the damn DSR and DCD and RI pins anyhow.

Early computers didn't have UARTs. The code would have to be in a function polling the bus to get the bits or they'd just disappear. The extra wires were to prevent that from happening by letting computers tell each other when they were ready to send and receive.

>> No.1951518

You can read and control the signals manually with ioctl(TIOCMSET) etc. Look at the source code for a comm package (e.g. minicom) if you want examples.

The tty driver doesn't use those lines, although it does use DCD if the CLOCAL flag isn't set. RedHat produced a patch to use those lines, but it never got merged.

>> No.1951521

Yeah, as I said, bitbanging. I also saw the redhat patch when I did look into it.
Bullshitting these externally was the path of least resistance.

>> No.1951704

unironically contact the manufacturer.

>> No.1951706

also are you the guy asking about resistors to lower the volume of a speaker thing? did that worked out well?

>> No.1951708

I say this because big medical shit like wheel chairs are usually low-volume and most of them are almost artisanal in making. Each manufacturer does everything differently so you really have to contact them. (send a email and call them).
I was giving minor help to one of the respirator projects that sprung up because of corona and there was zero information on the web about medical shit (much less about calibration in the case of sensors). We really had to call and email them to get the sort of data you asked for.

>> No.1951711

>respirator projects
How did that go in the end? IIRC they need pretty stringent safety interlocks of all sorts that would make getting a community-made respirator approved by some regulatory body exceedingly difficult.

>> No.1951717

New Thread Everyone

>> No.1951718
File: 52 KB, 720x834, olavo.jpg [View same] [iqdb] [saucenao] [google] [report]

It was a experimental setup made by a Brazilian university. Many were made and some quite competent. I lost touch with the project but I do not think it went past a prototype because the unfortunate stabbing survivor we call president cut the funding of all such initiatives (and most, if not all research desu).
I know that in Brazil a group made 3D printed consumables, used when you "tube" up a patient and had to be discarded after each use, free of charge for some hospitals. The regulatory agency must have looked the other way in that case I think.
After all investing in national tech is a very stupid thing to do since every student and researcher is a sex-addled, drink-and-drug sodden communist spies.

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