/diy/ - Do It Yourself

I've only ever done ultra basic soldering with a pos iron. I'm looking to make some speakers and was wondering if this is all the info i need for making these sort of circuits? Just pick the parts and solder them to a breadboard?

>>2513814
Yeah. Though for speaker crossovers I'd tend to go for tag-strips instead of protoboard. You have 5 nodes total, so a 5-tag one like pic related would be my go-to. The drivers should hopefully have polarity marked on them, but if not you can find out through trial and error.

>>2513851
Ive laid down some circuits hefore but im struggling to see how this eould work. Dont components need more space between them?

I'm going to build a solar tracker and when I took a look at some of the sensors available on the market I noticed that a lot of them use simple LEDs. I didn't even know LEDs could sense light before that, so now I'm curious why one would use these instead of photoresistors? Or even "photodiodes" "phototransistors".
Between all these, what would work best for a simple solar tracker made with an Arduino?

>>2513964
light trackng retarded. just use mathematics to calculate it based on the time, date and position (latitude/longitude) of the device. if you really want to do light tracking, put an inverted cone over the sensor. then literally rotate the thing around until you hit the sun, because your detection angle is limited, because of the cone. a simple led reflector/diffuser bizzo will work

>>2513968
>light trackng retarded. just use mathematics to calculate it based on the time, date and position (latitude/longitude) of the device
I know anon but I already answered why I'm not going this way on the other thread, see
>>2513265
>>2513282

>>2513968
>if you really want to do light tracking
>a simple led reflector/diffuser bizzo will work
not sure I can visualize what you're suggesting anon, sry I'm retarded
Most of the DIY sensors I've seen seem to use a cross divider between the 4 LDRs, but a lot of commercial ones don't even use any divider and they all seem to use LEDs, I'm wondering why they don't need any with LEDs. Are they more directional in the way they sense light compared to LDRs?
https://imgur.com/a/SZybJ47

>>2513810
i've built very simple SMPS in the past with ICs that integrate the switching FETs. desu there wasn't much of a challenge because i just followed the reference designs and it worked fine in practice. now i want to try making something more complex. i plan on starting small with a <50W buck converter. i happen to have an STM32G474 nucleo board on hand. is there any benefit to learning analog SMPS design (using an off the shelf SMPS PWM controllers) vs jumping straight into trying to build my first "real" SMPS with a microcontroller and a discrete time control system?

>>2513796
This one
It goes between the power input (the white connector with big wires going in) and the pci-e slot 12v+ pins
I don't have a way to draw on the image on this macro cam but it's the 6 pin marked 58AF just right of the big capacitor in the very center of the image.

It switches on a 75w gpu but the fet is rated for only 58w. It's usually fine because it averages at 50w under load
The problem is the vdroop is so bad, and the gpu will not sustain itself past 11.6v in, it will just shut off after a while, if not immediately.

>>2513988
Well now it's just left of the capacitor

>>2513988
It looks burnt (the "F" in 58AF), and the traces look a little crispy between the Parade IC and the passives leading to the FET.

>>2514003
he's doing the best he can ok

It's actually not, just the marking was scuffed from new, also in that pic all the heavy amperage is beneath it, none of the traces near the parade qfn are hurt, just a bit dusty

>>2513902
>Dont components need more space between them?

the solder tags are used to connect 2 or more points in a node.
how you arrange them in space is all up to you.
google ''point-to-point" wiring.
personally, i'd forego the terminal strip, and just use an air-breadboard approach.

>>2513964
So basically you have two photodiodes/LEDs separated by a small barrier that's normal to the solar panel. You can connect them to either input of a comparator and use the output of it to drive the motor. That way, if the voltage at one photodiode is greater than the voltage at the other, the motor will rotate them (and the solar panel) to lower the difference. A more stable approach would be to use an error amplifier (op-amp with a somewhat high gain, 20-40dB) that feeds a PWM circuit or MCU that controls the motor, that way it won't be giving the motor full-power when it's close to equilibrium.

>>2513988
I'd browse your electronics vendor of choice for a FET in the same package that can handle significantly more current at the same gate voltage. A factor of 2 would be what I consider safe, so using the Vgs/resistance graph and P = R*I^2, you'd want the resistance to be 4 times lower. If that graph doesn't exist, then you can probably estimate it with the Rds-on ratio, assuming the Vgs-threshold values are approximately the same.
If you can't find one, then I'd bolt a TO-220 onto the chassis with wires going to it.

>>2513973
LEDs have beam of typ. 120°, LDRs are flat meaning almost 180°. I think even LDRs would work without the divider if you angle them properly like they do it with LEDs. As to why LEDs work for this - it becomes clear when you realize that solar panels are also LEDs, if you put power into panel it will emit light (infra-red). Its all just PN junctions, but with various optimizations to increase efficiency.
https://www.youtube.com/watch?v=6WGKz2sUa0w

>>2514090
>significantly more current at the same gate voltage
Theres one a little bit better but not much, it's a super tiny footprint

>>2514109
Yeah. My personal design philosophy is to design FET circuits such that the chip would be fine in free air without significant PCB heat-sinking, if only because it makes thermal calculation easier.

I'd jumper it.

>>2513974
In my experience, you can get most SMPS ICs to work perfectly well if you follow the design guidelines. They are most of the time quite straightforward to use and you will seldom need to roll out your own solution. If you've used one on-line SMPS IC, you've used them all. If you want a "challenge" try making a circuit with less integrated, more general purpose part like UC384X PWM controller IC instead of a dedicated SMPS IC. Try to minimize ripple and make the output power cleaner.
If you want to learn how things truly work under the hood, the best thing is to take an off the shelf microcontroller and design an SMPS around it. Choose the switching devices, inductors (DIY? Or off the shelf?), gate driving solution, etc. Then comes the absolutely fun part of programming the MCU. You can use a number of different approaches to write the control loop. This part is the most enjoyable and i would highly advise you to use an oscilloscope to view various waveforms, they can reveal many things about what's happening.
I would also advise you to read about magnetic materials for a bit. You don't need to get a PhD, you just need to revise highschool stuff like magnetic permeability etc

>>2513974
Bust out the old TL494, connect its output to a decent half-bridge MOSFET gate driver IC, and have some fun. Can easily get hundreds of watts with that kind of setup. IIRC computer PSUs still sometimes use TL494-derived chips.

>>2514092
Pretty interesting, thanks.

Is there a standard pinout for a 4-pin logic voltage UART on a 1x04 connector? I'm not finding any standard ones, only a collection of different ones, but the hardware hacker guys seem to just know. I'll eventually make my own cable for it anyhow, but having that cable be useful for other things would be nice.

Also I realise I need a panel/chassis-mount connector for this. Nothing stopping me from just screwing a bit of perfboard with the pins on it to the inside of the 3D-printed case, but I'd prefer something nicer looking and more rigid.

Also another reminder for myself to not mix up RX and TX again.

>>2514169
>I need a panel/chassis-mount connector for this

apple used a serial 8-pin mini din once upon a time.
you can prob find a 4-pin s-video cable in your junk box, and can pull a female jack out of some old VCR/DVD or whatever.

i need to stop browsing /g/ which is just bullshit and start browsing /diy/ which is a fun board

And today i converted some shitty cold white lights to comfy warm white

>>2514180
>converted some shitty cold white lights to comfy warm white

last week i came upon a ceiling fixture where you can change color temp by moving a solder jumper on the PCB holding the LEDs.
this has lead me to think that maybe you can do that just be varying the series resistor or the number of LEDs in a series.
will have to trace it out one day when i'm really really bored.

>>2514169
>Also I realise I need a panel/chassis-mount connector for this. Nothing stopping me from just screwing a bit of perfboard with the pins on it to the inside of the 3D-printed case, but I'd prefer something nicer looking and more rigid.
M12 connector is the one used in industry for pretty much everything from sensors to ethernet

>>2514178
>you can prob find a 4-pin s-video cable in your junk box
Nah I'm a millennial. I've got some 180° 5-pin DIN sockets, but I bought them to use with an AC power brick's 3-pin split-rail output, and would rather leave not use the same connector for logic-level signalling as for a couple of 11.5VAC rails, in case they get mixed up on the benchtop.

As for things I can easily buy, I guess a 4-pin mini din (PS/2) would work. Honestly I think I'll 3D print a seperate connector block to be keyed into a gap in the main housing, that way I can swap it out for any connector I want, including a USB B socket with integrated USB-to-UART converter IC. Or even a TRRS socket, considering that's what my pinephone can use to speak UART.

>>2514202
I was more wanting a connector that's standard for UART specifically, though I guess things either just use DB9 with RS232, or they have an internal header. Neither of which applies to me (no I'm not cramming a MAX232 in here). Not really any demand for external UART, but I think I'll be using this somewhat often.

Also is that a screw-on connector? Would rather have something a bit easier to connect and disconnect.

>>2514111
I can't jumper it. What the fuck.
My idea is to take it off and solder on wires to a free floating fet or cut the trace on the other board right before the gpu and use the fet 12v to power up a solid state relay which simply connects V+ to the other side of The cut trace

>>2514355
>My idea is to take it off and solder on wires to a free floating fet
That's what I meant, having jumper wires going to an off-board FET clamped to a chassis or whatever.

SSRs can be a bit slow to turn on and off.

>>2514362
>SSRs

>>2514364
Based.

>create H bridge PCB with nice 50A MOSFETs and gate driver ICs
>connect dev board to switch at 60Hz
>connect H bridge board to small 16V to 120V transfomer
>turn on the system, works fine, I get ~120V on the secondary

>connected board to primary of massive 10A 12V to 220V transformer
>turn on the system
>get a single flash and pop on the main MCU on the dev board
>MCU gets a hole burnt into its package
I got fucked by back EMF didn't I? I knew I should've added a snubber or whatever

>>2513810
Any electricians here? Am I allowed to run two 3-conductor 250MCM cables inside 1 conduit (3.5" conduit)? I was told it's not allowed, but I looked through the code book and it says you can fit 6 conductors inside a 3.5" conduit for an RWU90 cable. As far as I know, a 6 conductor cable is the same as 2 insulated 3-conductor cables...is that right or wrong?

I'm following the Canadian Electrical Code but NEC is ok too

>>2514375
Of course you can. The question is how much amps are you going to push trough cables? You need to apply appropriate reduction factors depending on installation method and number of circuits inside a conduit (pic related - https://www.legrandgroup.com/sites/default/files/Documents_PDF_Legrand/Solutions/Power_guide/Power-Guide-Legrand_EX29008.pdf ). After you determine those then you choose appropriate circuit breaker, check that you didn't exceed maximum permissible length and then adjust cable size if you don't have enough amps. In your instance if you have two single phase 3 conductor cables this means you have 2 circuits - total current carrying capacity is reduced by 0.8. Then if you look at another table on page 20 and if I assume you'll be installing this according to method B1 (on/inside masonry wall) then you get 249A, reduce that by 0.8 = 199A. From thermal design point of view cables and insulated conductors are the same, but I don't have access to standards to check if there is some specific directive about conduits and what can go inside.

t. not electrician

Another one done
>>2514183
Connect a switch to it

If I have a linear bench supply that runs on 110V, is it possible to put a large inductor in series with the power supply transformer's primary, forming an inductive divider, so that the bench PSU can run on 220V?

>>2514560
What the fuck

>>2514560
Anon we need the model, I have never heard of a linear power supply running straight from 110V. The heatsink must be huge

is there any rule of thumb when replacing a capacitor? I want to replace a capacitor but I only have one that has 15% less capacitance. How much is the range i can play with? is that too much?

>>2514599
What matters the most regarding safety is the voltage rating and type of capacitor. For polarized caps as long as you're at the same voltage rating or higher it's good, capacitance can vary by 20% so for filtering you don't need an exact value. For caps that connect stages together capacitance does matter a lot since that also affects the attenuation of the signal depending on the frequency.
For unpolarised caps, what matters most is the voltage rating and class. Don't replace a cap with a class lower than the original one, ESPECIALLY when dealing with main voltage.
There's more to it but this is the most basic stuff, there's also temperature ratings, leakage current, ESR, ESL, etc but that (usually) doesn't matter.

>>2514616
Forgot to say that for capacive droppers and oscillators the capacitance matters a lot, so keep that in mind.

>>2514616
>>2514617
ok, thank you for the explanation. It's a filter capacitor, so I think I'll be safe. Cheers

Gonna sound retarded, but I was going to wire some stuff together when I realized I forgot to grab connectors. I didn't feel like grabbing any so here's my ghetto connectors made from a soup can and a pair of tin snips. My plan is just to clamp the folded edges over a wire with a pliers

>>2514212
>Also is that a screw-on connector? Would rather have something a bit easier to connect and disconnect.
Yes those are screwable, waterproof and theres a lot of different variants. But of course those are pretty heavy duty and something like db9 or mini-din is more neat depending on use case.
>I was more wanting a connector that's standard for UART specifically
I googled what UART means and I understood it's just abbreviation for serial buses in general. What you want to do is check voltage level on your bus and is it differential communication like RS-485, then check if any RS standard is similar. All those RS standards have pinouts for DB9 connector.

If there's no standard for your UART, then create your own. If you want to make sure you don't mix it up with rs-232 and fry something, I would use the "ring indicator" and such pins which probably haven't been used past 20 years or so

>>2514635
Those are abiko connectors you made? Just tighten the wire around screw without connector as those makeshift connectors look like sorry but not really reliable unless you solder them and also the metal on tin can can cause corrosion

>>2514635

such beauty and ingenuity.... i'm left speechless.

>>2514371
Bit of an odd situation. In general I'd just avoid feeding square waves into mains-frequency transformers because they can saturate from those harmonics, but I can't think of any situation where a failure kills anything other than the FETs. Inductive spikes shouldn't be an issue since one FET should always be conducting, but after one or both FETs are dead, it's possible that them (or the power rail fuse) going open-circuit could promote a voltage spike.

>>2514560
No, because a fixed inductance means a fixed impedance, so voltage drop will be a function of current draw. Best option would be to replace the transformer. If it's a fancy multi-tap jobbie then you may find it more cost effective to rewind it. If you're real lucky then it already has two primary windings in order to be rewired for 120V or 240V.

>>2514595
Mate he said transformer.

>>2514648
I say UART because it's the term MCUs use for their serial hardware that's functionally compatible with RS232, but at CMOS logic levels. I've got optocouplers on my output stage I could probably go up to 0-12V signalling, but it has to be self-powered. Any standard without its own DC power and ground lines won't suffice.
The only RS serial standards I can find either have strange logic levels or do differential signalling. So I'm making my own I guess.

Serial usually has male-to-female cables, right? Makes the whole swapping RX and TX thing less of an issue. So whatever connector I pick should have chassis-mount and cable-mount versions of both the male and female. I see why DB9 was so popular. Mini XLR, 3-pin + shield or 4-pin? Using two different connectors works too, like 180° and 240° 5-pin DINs, or BT and 4P4C. Not USB A and B though, would get too confusing.

Or maybe I just resign to never connecting two "data terminals" together?

>>2514763
If you want to make it standard so badly, why not just get some driver chip like the max232 to convert it into standard bus like rs-232 or rs-485?

>>2514775
As I said, I've got optocouplers in order to seperate the voltage lines of this circuit (mains thermocouple controller) from the comms lines. Pretty sure I couldn't self-power a MAX232 off the incoming signal voltage(s).

>>2514781
There are isolated drivers which don't need any extra optocouplers and such. See for example this https://fi.farnell.com/en-FI/analog-devices/adm3061earz/rs485-transceiver-isolated-85deg/dp/4023956

>>2514781
>>2514781
There are isolated drivers which don't need any extra optocouplers and such. I hope this is the correct link now https://fi.farnell.com/en-FI/nve/il3585e/isolated-profibus-interface-3585/dp/1139045

finally got my boards
now I have to dig out all the components
it's gonna take hours

How do you select parts? Sites like digikey have extensive filtering but I'm still lost.

>>2514808
Post boards, explain projects.

>>2514809
Sorting by price and availability gives some degree of further selection. Then just go through datasheets and figure out which ones have the best specs for their price, or are otherwise more versatile. Good documentation like appnotes and design tools is also a favourable feature.

Versatility is especially important for hobbyists, as getting a stockpile of parts to design projects around is a lot cheaper than making a small order for each project.

>>2514809
Sorting by popularity or viewing best sellers (Farnell) is pretty handy sometimes

Has anyone tried ChatGPT yet? I wonder if it can draw schematics or answer EE questions. Don't want to give them my phone unless absolutely necessary.

>>2513810
Brainlet here; I need your help. I did look this up, but I can't find a straight answer. I'm trying to read resistance with my multimeter. I have my probes on hooked up, and it is reading 2.6 on the 20k ohm setting. Does that read as a straight 2.6 ohms, 260 ohms, 2600 ohms, or 26k ohms? if its a simple 2.6 ohms, why doesn't it read on the 200 ohm setting? I DON'T UNDERSTAND
also, for some reason im able to get the same numbers even with the leads not in COM

>>2514889
The 20k setting means maximum of 20kΩ, so it definitely won't be 26kΩ. Assuming it's a 2,000 count meter and not a 20,000 count meter, if the entire display reads 2000 then that's 20.00kΩ. If it reads 2.60 then that's 2.6kΩ, which I imagine is what your situation is. No reason for the decimal to be anywhere else, having it describe kΩ makes the most intuitive sense. I imagine it won't measure when set to the 2kΩ setting? That means it's definitely above 2kΩ, so then there's only one thing it can be.

>for some reason im able to get the same numbers even with the leads not in COM
The 10A socket or equivalent is directly connected to COM via a 0.01Ω sense resistor (give or take an order of magnitude) so it won't make a significant difference to your measurements. Leaving a probe there is bad practice though.

>>2514889
It's 2.6k Ohms. The meter blanks out when the value is too far out of range because it's not an auto-ranging meter.

>>2514893
>>2514894
Thanks, bros. That really lays it out for me with the other readings too. I appreciate you :*

>>2514595
>>2514562
Pic related, 630V - 0.11A output. I'm planning to use it to charge these capacitor(450V/4700uF) for some fun. I have like a dozen of these high voltage capacitors

>>2514763
Yeah, I'll see if it can be rewired for 220V. If not I can just always get a cheap 220-110V step down transformer.

>>2514987
If you want to charge one of those caps, just rectify mains and use a series power resistor. 220VAC peaks at 310V, so it's getting close to fully charged.

>>2514993
Nar, did some quick math using E = 0.5 x C x U^2
>310V I get 226 J
>450V I get 476 J
So using mains I can get like close to 50% charge of these caps

>>2515008
Then double diode rectify it to 620VDC, then use zeners (or some sort of TL431 circuit) to keep it below 470V.

>>2515012
I mean I COULD, but why bother when I have a proper HV adjustable power supply already that just needs a simple requiring or just a cheap 110V step down transformer to use?

Say I wanted to take the ready buck converter schematic from this website for the TPS62932 IC and turn it to a PCB + add a trimmer pot to adjust the output voltage.

How the fuck can I do that without having to complete a master's course in EE?

>>2515077
Reconstruct the reference circuit in KiCAD's EEschema, with your choice of passive package. You see those two resistors from the output to the feedback pin to ground? Replace them with the trimpot, wiper at the feedback pin. Then import that into PCBlayout and shuffle components about for a bit. Watch a video on PCB layout tips (avoid altium shills) before you start, basics are:
>don't put shit stupidly close together, or stupidly far apart
>obey the DRC checker
>grid align everything
>minimise switching loop area first
>minimise other trace lengths within reason
>ground planes are good
>don't forget your input and output connectors and board mounting hardware
Check the datasheet for any recommended PCB layout while you're at it, chances are half the work is done for you. Post here before getting it fabbed so us friendly fucks can redirect any mistakes.

>>2515079
I did a preliminary check for where I can get all the required components and the only place I can find it where it isn't on backorder until 2050 is from this supplier on Alibaba:

https://www.alibaba.com/product-detail/Good-price-TPS62932DRLR-IC-chip-Buck_1600626903806.html

Do they actually sell you the genuine part or some chinkshit under performing knock off? Normally I wouldn't really care but the whole point of me diving into this is to actually use the IC that can actually deliver the promised 98%+ efficiency.

Also, as for the passive components, I assume just going on Aliexpress and getting the cheapest shit possible won't really help with the efficiency I am after, right?

>>2515080
Don't buy specialised modern ICs from noname chinese sellers. They're either fakes or soldering bin scavenges, avoid in either case. Start off by choosing a part that's not out of stock next time. Check the "in-stock" button on your supplier of choice (ideally one with a local warehouse) and filter by your desired specs.

Ali passives aren't terrible, though if you're already making a digikey order I'd get passives from them. That way you know the tolerance, voltage rating, tempco, etc. for future projects. I've got chinky ceramic caps and I'm tempting fate somewhat by putting the 10µF 0805s on a 12V circuit.

>>2515083
okay so I did find a IC that is actually available from DigiKey:

https://www.digikey.gr/en/products/detail/texas-instruments/TPS563300DRLR/16669318

and the relative schematic is pic related.

should I care about any of the options on the left?

if not, do I just go to town with this schematic and start designing?

I should also note that this board's purpose will be to drive a small, brushed DC motor that has a small gearbox on its shaft - don't ask me why I need a highly efficiency buck converter for this job instead of throwing a random LM2596 board that I have on hand, so I don't know if that changes anything.

>>2515087
Yeah it's probably fine to just start with that. I'd be sure to read the datasheet though to ensure all the specs are acceptable, like quiescent current and efficiency at certain operating points. Also do the thermal calculation to see what temperature it will be running at, and how much PCB heat-sinking you'll want. Chances are the datasheet's layout example will be good enough, if you're at liberty to put a bunch of vias around the IC.

Also consider what will happen if the motor is still rotating when the power cuts out, where its voltage may go to.

>>2515091
currently a relay after the buck converter controls the operation of the motor

do I need any special provisions for reverse current flow? maybe a diode or something?

I mean, this is what I am using right now:

https://www.aliexpress.com/item/32907694807.html

I think it only has 2 diodes in the input and output to protect it from reverse voltage, but I could be wrong

Absolute newbie regarding electronics, but for a small project I want to run some small string lights from lithium batteries (friend gave me some disposable vapes kek) with a small solar panel charger.

Is this possible?

>>2515131
well yes but will it be worth it?

>>2513964
What's next? you'll be sanding off transistors to turn them into photodiodes.

>>2514180
Greeting to /csg/'s Mexican femboy.

>>2515188
lol I've read of boomers doing just that while researching this topic

>>2515087
am I doing it right?

Anyone able to help a total noob? I want to DIY a directional antenna to use with a phone signal amplifier, where do I start?

absolutely retarded question, but i'm trying to make a buck converter in lt spice. why isn't this working? i realize it's open loop configuration but shouldn't i be seeing some sort of pwm on the output? what am i fucking up?

>>2515314
>i'm trying to make a buck converter

>>2515234
No, RP1 should span from the output to GND, with the wiper (the arrow) going to FB.
Is that the custom symbol? Kinda barebones, the pins may be backwards, and people usually ignore the actual order of the pins and arrange them for convenience. Say Vcc on top, GND on the bottom, inputs on the left, and outputs on the right. FB below SW below BST. NCs hidden.

>>2515314
Default NMOS is ass, set it to an actual transistor before trying again. Also since you're making a top-side converter instead of a bottom-side one, the feedback signal would go up with decreasing voltage, not down. Either way follow the datasheet.

>>2515339
>the feedback signal would go up with decreasing voltage, not down.
i just redrew it with a PMOS in the more traditional >>2515337 layout, still not working.
>Either way follow the datasheet.
data sheet doesn't have an example of a buck converter and i'm learning this shit for the first time.

>>2515339
I have (almost) no idea what I am doing at this point desu

>>2515346
PMOS FETs turn on with a negative gate-source voltage. You're better off trying to run it with a low-side switch, or as an inverting boost converter. A high-side switch with a high-side floating driver works too.

>>2515347
Here. You were getting close, if you ignore the shorted output caps. As I said, it's more intuitive when you arrange the custom symbol such that BST is above SW, which is above FB.
What's up with the three 820nF caps?

>>2513810
Has anyone ever tried to convert a 512x384 greyscale CRT from a classic Mac, specifically a Macintosh Classic, to accept 640x480 VGA? Plenty of documentation on modifying a COLOR Classic to accept VGA but nothing about any of the greyscale compact macs. Is my best bet just trying to find a burnt out 9” b&w monitor and swapping out the tube driver electronics?

What can cause a powered speaker to randomly start hammering the woofer at maximum load? No audio sauce but the thing sounds like a M249 being shot in my loungeroom when it does it. The woofer is also visibly moving like a good inch forward and back. Id love to trouble shoot it but i feel id probably end up killing myself on a cap

pulled this out of an old broken piece of hardware that had a failed sensor. What is my first step in re-using this LCD?

>>2515583
Tube driving should be relatively universal amongst CRTs. As for the signal side, you'll just need a composite timing chip like the LM1881. But that stuff is probably all of what's implied by the colour swap, so maybe there's something more specific that's troubling you.

>>2515698
Figure out the pinout, and remount it to a board of some sort. You may have to probe it with a logic analyser while it's running, or you might be lucky and find that it's easy to narrow down what display it is and hence get a datasheet.

>>2515698
Did you try unplugging it for 60 seconds and plugging it back in?

>>2515703
gotcha, I will try to dig up my logic analyzer later in the week and try to see what I can find out. are there a ton of different LCDs out there that vary from model to model?

>>2515704
haha oh yea i did, and more. got it working again fora second then it just would not stop giving errors no matter what i did. think water got into the sensor and shorted it rip

>>2515716
>are there a ton of different LCDs out there that vary from model to model?
Yeah, but there's less variation when it comes to the display drivers. The same driver chips are used for a fair few different displays out there.

I have no clue about microphones, what differentiates a cheap dynamic microphone from an expensive one. I looked up some cheap vs expensive videos on youtube and it seems like the cheap ones have more noise and more dull sounds.

More dull makes sense if for example the membranes can't pick up sounds that well (my guess) but what causes the noise? Couldn't one just fix the noise issue with a ferrite core or something?

>>2515558
it says to use 3 of those for some reason - wouldn't the correct thing be to use a single, equivalent cap?

>>2515738
No. The reason is all capacitors are inductors and different types and values will have different self-resonant frequencies. A mix of capacitors is usually desirable.

>>2515732
Dynamic microphones fundamentally have an open-core inductor at their heart, so they're prone to EMI. Could try to remove this by having two in a humbucker configuration, which I think would require having the magnet of one being backwards compared to the other.
Frequency response is a function of membrane mass, as the membrane acts like a mass on a spring. It acts like a low-pass filter, so high-frequencies will be harder for it to pick up. Depending on the Q factor and resonant peak, there may also be a nasty spike or two somewhere.

I'm a condenser mic kinda guy, condenser membranes can be orders of magnitude lighter for the same area. The requirement of a low-noise moderately high bias voltage is a bonus project. That or an RF microphone, which are cool.

>>2515738
It's more likely to mean 3 in parallel, not in series. Adding caps in series is bonkers unless they have really low voltage ratings.
The datasheet doesn't say anything about 3 caps being used, I'd just go through the section that says "8.2.2.6 Output Capacitor Selection":
>https://www.ti.com/lit/ds/symlink/tps563300.pdf
And see what capacitance you get.

>>2515742
Yeah but 820nF is a bit too large to be getting away from low self-resonance frequencies, and you'd want to have a combination of different values.

>>2515690
>What can cause a powered speaker to randomly start hammering the woofer

amp is oscillating due to one of a thousand reasons, usu components with AIDS.

>id probably end up killing myself on a cap

you've been brainwashed by weenies.
voltages are prob under 30V so it's no more dangerous than handling a 9V battery.

>>2515698
>re-using this LCD?

ignore the logic analyzer advice.
multiplexed LCDs dont use logic levels.
usu they have multiple backplanes and the segments are fed 4 diff voltage levels.
it's possible to reverse engineer with a scope, but not worth the hassle.
stick to LEDs and be content in your caste.

>>2515583
>swapping out the tube driver electronics?

that sounds a bit crazy.
you shouldnt have to mess with the monitor much.
a saner approach would be to convert VGA signal to something the monitor can recognize, like composite.
see pic.

I have an arduino with an ethernet shield, and I have my little program working on it. Now I need to communicate with it from a PC over ethernet, using TCP.

Anyone know a simple way to send a couple of arbitrary bytes to a known IP/subnet/gateway/Mac? And read what comes back?

Is there a way to do that from the command line in windows? As I understand it, telnet is close but not quite what I'm looking for.

>>2515937
Download a terminal client like puTTY and use ssh or serial.

I need to solder a button onto some of those cables that connect to headers like on a computer motherboard or a mc like an arduino or an RPi.
What's the best way to go about this? I have jumper wires a plenty I could cannibalize connectors from, but from experience I know those aren't really solderable. I don't have anything uncrimped and my shithole country doesn't have any stores that would have them, I'd have to order from China and wait hope I get them before next summer..

>>2515943
Ok, great, thank you, this appears to be exactly what I need. I've set up putty, and hooked everything up and I can set it to raw mode and send my bytes over, which almost works as expected, but it seems like something extra is being sent over? An end byte maybe?

When I have it hooked up to a second arduino with ethernet shield programmed to just send bytes via tcp, it doesn't exhibit the same behavior. From putty though, it gets it's expected two bytes, executes the action as normal, then starts the next loop and reads two 4-bit numbers from the client (puTTY) that don't display as ascii characters.

>>2515698
>hotbar attached LCD
>glop top chip
its trash anon, unless you want to reverse engineer each segment's connections and use zebra strips and custom boards

>>2516100
Create a new session and save it. In the left pane, click Window > Translation. Select CP437, ISO-8859-1, or Windows-1252 as the Remote character set. Save the session, connect and try again.

>>2516100
>>2516136
or use Realterm instead.
https://realterm.sourceforge.io/

I know that batteries while able to store more energy than capacitors can't give out the demanded power out as quickly so how much power is ready on demand in batteries percentage wise?

>>2515937
netcat
If you don't want to configure your OS IP config to communicate with the device, there are hacker/security tools that can build and send raw packets to a network interface.

>>2516144
Depends on the battery, and it depends on what sort of timescale you’re talking about. 100C lipos can fully discharge themselves in 36 seconds. I doubt any modern battery (except for maybe SLAs) have any significant chemical delay since the electrodes are so tightly packed together, so it’s just ESR that’s the limiting factor. Current is limited by battery ESR, so you should be able to figure out maximum current via Ohm’s law.

>>2516136
>>2516140
>>2516145

Awesome, thanks guys. I've got it functioning with a placeholder static up. Now my problem is that the device needs to use a private/internal IP address, but when I program it to use an IP address in that range {172.16.0.0 - 172.31.255.255}, I can't ping it or anything. Honestly not sure what I need to do to make it respond. I'm just using a linksys wifi router as a wired network switch and it's just 192.168.1.1

Do i need a switch with a similar 172 address for this to work?

>>2516161
Your router's firewall or the WIndows firewall might be blocking it. Need more info on your setup. What is connected and how? Upload a pic or drawing.

>>2516161
Is your router configured to route it?

Since this involves electrical stuff and the stupid question thread is on autosage I'll ask here. How do LED tubes factor into the maximum rated wattage for a given light fixture? If pic related is rated for 32w T8 bulbs, I assume something rated 28w would work? The ones I've got in right now are 32w "equivalent" while supposedly only using like 18 or something. They're alright, but I feel like they could be brighter. I'm using them as grow lamps for houseplants. Most are doing OK but my Echeverias are getting a bit stretched. Also, would 40w t8s work since it will apparently also fit 40w t12s?

Come to think of it, I'm honestly not sure if the wattage ratings on fluorescent lamps are strict rules or more "guidelines" like with screw in bulbs (i.e don't stick the 300w space heater lamp in the 1950s 40w 2 wire fixture)

>>2516337
The actual wattage of the replacement is likely all that matters. So you should be able to go up a step or two.

anybody know a good replacement for these old jvc 8 pin connector fuckers that break easy?

>>2516161
>wifi router as a wired network switch and it's just 192.168.1.1

your router may not be smart enough to recognize 172.16.0.0/etc as an internal address
try the other options
10.0.0.0 - 10.255.255.255
192.168.1.2 - 192.168.1.255
the last one is sure to work (as long as there's no collision) coz all your network gadgets are assigned that range.

why does touching black wire on computer PSU to the metal casing of the psu produce large spark and make the psu turn off?
i thought all grounds are connected

>>2516379
>good replacement

solder is the universal connector, that neither time nor storm can defeat.

>>2516424
>thought all grounds are connected

your supply is the exception that proves the rule.

>>2513810
What is the best way to cope with leakage inductance in a flyback converter. I refuse to add a flyback diode, I don't want to piss away energy. I have a bunch of 600V MOFETs so hopefully they won't blow.
Also what is a good way to measure inductance without selling my kidney and buying an LCR meter. I have bought a few E-I cores and would like to wind my own transformers.

I have massive respect for you power electronics guys. I'm originally a 1.8V, 3.3V and 5V guy but I've been playing with power electronics and its very fun

>>2516432
>good way to measure inductance

there's prob 1000 youtube videos on the subject.
pick one that meets your collection of instruments.

>selling my kidney and buying an LCR meter.

alibaby has 'em for $12 or so.
$30 for the the super-duper ones.

>>2515745
>Dynamic microphones fundamentally have an open-core inductor at their heart, so they're prone to EMI
My house has extremely bad interference for whatever reason, so that makes a lot of sense. I know this because I was messing around with a DIY AM radio basically with just coils and capacitors and managed to pick up an Arab or Turkish radio one time, but for the most part it was just extreme interference the majority of the time.
>I'm a condenser mic kinda guy, condenser membranes can be orders of magnitude lighter for the same area.
I don't have a proper studio for a condenser microphone so I can't really appreciate condenser microphones. I do have a cheap one but it's so sensitive that it picks up every little sound in the room, which is the point of them I guess, but I don't need or want that for just voice recording. I should have bought a dynamic microphone instead of the cheap condenser I bought because the main purpose I bought it for was voice chat, and I had no clue at the time that a dynamic microphone like the Fifine K6 would have been the far superior choice in my case and even cheaper.

Thanks for your input, I'll mess around with some cheap dynamic microphones.

>>2516438
>My house has extremely bad interference

seems to me a good way to minimize infeterance is to fit an amp (and maybe a DAC) right inside the mic handle.
instead of having a long cable.
like USB mics do.
i have a semi-pro one that takes an AA cell right in the handle to power an amp.
a pedo priest gave it to me free coz he said it stopped working.
i asked if he had checked the battery and he looked at me like i was crazy.
mics dont use batteries; unlike his MyLittlePony vibrators.

is there a way to resurrect a dead IBM Laptop (Thinkpad R50)?

some time ago a friend of mine gave me this old ass laptop with a broken screen (50% of the screen didnt work), it wasn't a problem, i used it as a server, and if i really needed to read something from the screen i hooked it up on my monitor. But one day, after being off for a while, when i tried to turn it on it refused to come back to life, none of the leds lighted up, the fan didnt start spinning, etc; it was dead.

I think it is some problem with the power supply, is there a way to charge its battery without it being connected to the laptop?

>>2516521
Buy a second hand ThinkPad charger. They're everywhere for next to nothing. I use one as a power brick for my soldering iron. The battery is almost certainly dead and then recharging it will entail cracking open the plastic case, resetting the self protection circuitry and applying 4.2v to the 18650 cells individually while constantly monitoring them to make sure they don't burn down your house.

>>2516521
Use the power brick on something else to rule it out, then check the DC power jack.

>>2516529
>>2516534
the power brick works, i think that the problem is the dc jack or something related to that.
anyway, how do i manually charge the battery detached from the laptop?

>>2514180
nice scorch marks

I figure this is the best place to ask, how the fuck are you supposed to resolder those? i have around 50 screens that are perfectly fine except those cables are always making a bad connection.

>>2516558
Forgot photo.

>>2516558
>>2516559
T-tips

>>2516566
Cant i just use a knife tip over the part that has problems?

looking for a power supply for DIY synthesizer builds, i was recommended to get a dual ouput power supply but the most ive found are expensive as hell. i found this for a reasonable price and was wondering if it would be a good fit for me

>>2516578
>Cant i just use a knife tip

whichever tip you use will cause the flexPCB to melt, leaving it irretrievably ruined.
once saw video of a viet guy fix one using an entire lab full of specialist tools and chemicals.

>>2516587
>i was recommended to get a dual ouput power supply

jeez louise, all you need is something like this (pic) you can make for $10, assuming you already have a transformer from some obsolete gadget.
you prob dont even have to make it adjustable. +-15V will cover 90% of your needs.

>>2516578
>Cant i just use a knife tip

whichever tip you use will cause the flexPCB to melt, leaving it irretrievably ruined.
once saw video of a viet guy fix one using an entire lab full of specialist tools and chemicals.
(i've had a small modicum of success using rubbery insulation stuff to press against the connector and the back of the case.)

>>2516587
>i was recommended to get a dual ouput power supply

jeez louise, all you need is something like this (pic) you can make for $10, assuming you already have a transformer from some obsolete gadget.
you prob dont even have to make it adjustable. +-15V will cover 90% of your needs.

>>2516547
If you don't know then you shouldn't be doing it. Just fix the DC jack.

i have a pcb with octocoupler on it, when power is applied to the diode on the octo it turns on
i need to reverse this logic, when i apply power i need the octo diode to be OFF and when i stop applying power i need it ON
what is the absolute easiest way to do this? I want to use as few components as possible since i have to deadbug them to the dumb pcb

>>2516606
i could be wrong but a well made power supply would be needed for constituency in sound output of the synthesizer

>>2516624
Start by removing the coupled octopi from your PCB
>serious answer
You can't drive the diode without current, if your signal is low voltage you could use an inverter or inverting op amp, but since it's an opto I'm assuming HV on diode side.
What does the circuit do?

>>2516626
>a well made power supply would be needed

the diagram shows an extremely well designed supply.
cant say the same that if you're buying some generic chink or indian box with no pedigree.

Is there a buck converter or some sort of power supply that can supply an output voltage between something like 50 mV and 500 mV? Current consumption will be almost negligible.

Also, it has to be small in size.

>>2516626
>a well made power supply would be needed for constituency in sound

that diagram is very well designed.
cant same the same of some generic chink or indian box with no pedigree.
also, sound quality is only marginally affected by the power supply, unless it was very poorly designed.

>>2516632
What are you driving?

Spent some Christmas bonus money from work on getting my basic electrics lab set up.
Already found a project to repair a collectable Spiro pinball machine I found in the attic that doesn't have sound.

Also really interested in the ESP32/8266 stuff. Though I'm not sure what board to get if I was mostly planning on using it for remote remote I/O projects with relays and sensors.

>>2516632
https://www.mouser.com/ProductDetail/Maxim-Integrated/MAX38912ATA%2bT?qs=Rp5uXu7WBW9ZSGyZzp1Vlg%3D%3D

>>2516632
1.8V LDO and a voltage divider

Are BD439/BD440 a good substitution for BD139/BD140?

>>2516641
It's a good start, Anon. The pipe is a nice touch, but that hat is no good. You need one of those mesh-backed foam hats with a misprinted WWF logo on it, and a giant foam Hulkamania finger for decoration.

>>2516656
>Hulkamania
Honkamania

>>2516432
TVS diode to protect the FET, snubber to suppress the spike. You have an air-gap in your coil, right?

>>2516438
Sensitivity is just a question of gain, the type of microphone shouldn’t make a difference. As the other anon says, put the amplifier as close to the transducer as possible. A grounded metal cage shielding around your mic is definitely something to look into.

>>2516624
Post schematic. If you’re lucky you can just flip the orientation of the LED (e.g. from pull-up resistor to pull-down resistor). Same applies to the photodiode.

>>2516632
RRIO Op-amp configured as a buffer, and a potentiometer. Put a transistor after it if you need higher current than the 100mA or so an op-amp can source. Just using a potentiometer also works in the event it’s a very low current drain. For consistency you’d want the pot to be powered from a stable rail, like from a linear regulator or voltage reference.

>>2516668
>>2516627
would this work? i need to reverse the fet logic

i changed the r2 from gate pull down to pull up and the opto instead of passing 12V to the fet gate now instead pull it to the ground, which should reverse the logic as i need
opto on = fet off, opto off = fet on

>>2515314
okay i got this working despite using n channel as high side without bootstrapping. my goal is a CC/CV power supply that can be used to charge single cell LiFePO4 batteries, 3.7V and 2A maximum from a 12V input. i want to control the CC/CV set points via MCU's DAC output. ignoring component selections, compensation (i haven't even looked at freq. response), etc., is this how it's typically done? i searched for reference designs but couldn't find anything. this appears to work, at least for a few different load resistor values and commanded current and commanded voltage values.

>>2514763
>Mate he said transformer.
Ah sorry, I misread it as a linear power supply running straight from 110V, with only a rectifier and filtering.

>>2516655
Yes but be aware that their gain is higher than that of BD139/140. They can also work up to higher frequencies.

>>2516680
Yeah.

>>2516747
It is a bit odd how "linear supply" somehow implies AC transformer-based.

>>2516430
>solder is the universal connector
true but the way this thing is built would make it hard to do any future work on it after that

Are mods like these difficult? Switching usb port to usb-c?

My kobo mini has a micro usb port which is getting loose. Would love to mod it so it can use ucb c (not for power delivery but just make cable managing easier).
https://www.sparkfun.com/products/15100

>>2516587
I bought me one of these..I like it, but it won't let me dial in like 0.005 amps, can't seem to set it lower than around 0.200 amps
https://www.amazon.ca/gp/product/B0B1HMZ1KV/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1

>>2516802
Where does the existing connector break?

>>2516803
Isn't USB-C even more delicate than micro-B? I think I'd rather put contacts on the back for a magnetic pogo-pin charging base.

>>2516826
the small part you push on that holds the wires down broke. There's no way to get under that board without taking the board out, so if I solder those wires, I wouldn't be able to work on it easily anymore

>>2516868
If there isn't a better connector with the same pitch, you could solder a wire on and add an inline connector.

>>2516872
hmm maybe

>>2516685
I know you are probably just doing this for fun, non practical purposes, and I probably know less than you about electronics, but I think the reason why you don't find anything is because BMS's are the only option if you want your batteries to output more than 3.2v, and all BMS's have a recharge function which has a over charge protection function (as well as other protection features).
Also I know basically nothing about electronics, but this stackoverflow post might help you with getting a lifope4 safe charging IC (they mention "CN3058E or MCP73123"):
https://electronics.stackexchange.com/questions/524634/charging-circuit-for-a-lifepo4-cell

>>2516883
yes this is pretty much just an autism project. i have some old 18650s laying around and i always wanted to build my own CC/CV power supply, so i figured i would start small. i actually found that post a week ago when i was looking at LiFePO4 specific BMS solutions -- there aren't that many on the market, which is surprising considering how ubiquitous they are in modern cordless power tools.

>>2516900
Don't cordless tools just use conventional NMC lithium ion? I've only ever seen one thing in person that uses LiFePO4s, which was a wifi doorbell for some reason.

>>2516883
>>2516900
like i said, i THINK this will work, at least on simulation. my biggest concern now is figuring out how to limit outrush current and then seeing if this will actually work IRL.

also i have no clue how to do bode analyis in ltspice. i tried following the official analog.com guide but my bode output was fucked up looking. :\

>>2516908
i'm pretty sure they just use standard 18650 cells, which are lifepo4.

>>2516910
Now measure power dissipation of the FET.

A bode plot only really applies to something with an input and output signal. I think what you want is to see the FFT.

>>2516914
Most 18650s are not LiFePO4.

>LiFePO4
We should change the name to something else like "thermite-free lithium" or "not white phosphorus".

>>2516915
i zoomed in on the "densest" part of the FET's power plot and it looks like an average of 3.95W. guess i'll need to do thermal calcs.

>Most 18650s are not LiFePO4.
damn i was completely wrong on that. i wonder what chemistry these no-name chink cells sitting on my desk are.

>>2516925
also i just discovered that holding down ctrl on an ltspice lets you pan around (with mouse) or zoom in/out (with scroll wheel.)

>>2516925
the shape defines the chemistry, if it's a flat square and very light, it's Li-Po (the battery in your phone).
if it's 18650 or 21700, it's Li-Ion (99% of the time).
if it's 38120 or 32650 or a big heavy square pack (which contains a bunch of cells), it's LifePo4 (it's not impossible to be Li-Ion though).
Also because LiFePo4 and Li-Ion have different voltages, you can just check the number in series and compare it to the voltage rating on the battery (Li-Ion uses 3s for 12v usually, Lifepo4 uses 4s)

trying to learn circuits

I have this more or less set up on a breadboard to press a button and blink the LED.

I'm trying to making a blinking LED with relay and capacitor only and without mechanical input

I've seen it done with one LED in these third world youtube videos but I can't quite figure out what they're doing to transfer it over to a breadboard

>>2516990
you need to gate the relay coil with one of its "normally closed" switches if you want it to flash

also for god's sake please use a freewheeling diode

>>2516990
>a blinking LED with relay and capacitor only

never heard that idea before.
you can get a relay to buzz (and flash an led) with this circuit.
but dont run it too long coz relays wear out.
to slow it down, cut the wire at point X and add a small resistor in series, like 22 ohms.
then add a big cap in parallel with relay coil, like 2200uF.

>>2516641
nice harbor freight bench
welcome to the club bro
(picture is pretty old, bench is a mess now)

>>2516990
I extracted a circuit diagram of one of those videos

I don't understand how the fuck this works
https://www.youtube.com/watch?v=-607kl-bE1Q

>>2517009
ok that would be a separate diode to stop current coming from the inductor, so like this?

>>2517026
Nice chair, where did you get it?
The desk is great except there's no place for your knees.

>>2517055
some computer chair I've had for a decade or so
think it came from walmart
it's honestly too short for the bench
the knee part does suck though
especially since my bottom shelf is full of stuff now

>>2517040
yes

I plan on getting this power supply for fast charging my ebike battery (12s2p lifepo4 3.2v), I was thinking of using this but I have a feeling the sound of a fan & and the house fire / dangerous aesthetic is going to annoy some shop owner if I used it for charging the battery.
https://www.amazon.com/gp/product/B0756B3P6M
I also have a laptop charger + boost converter I plan on using for "slow charging" because using fast charging all the time probably isn't good for the batteries.
Would you expect it's possible to swap out the fan for a quite computer case fan or if there is a cheap and superior power 40v supply? (I haven't bought it yet).

>>2517065
Do the batteries just want a constant voltage and have their own internal charging current regulating circuitry? I'd check that before hooking them up to anything.

If I've got an audio opamp with 22k input impedance, what value volume potentiometer should go before it?I don't want to lose high (audio) frequencies.

>>2517071
Pot value shouldn't influence frequency cutoff, though gain might be strangely effected. For example, a 100k pot will be pulled significantly around by that 22k resistor, changing the linearity curve of the pot. A lower value pot will mean less of this, but conversely it will pull around the source signal more. So how low a pot you can get away with depends entirely on the source. I wouldn't put a 1k after a high-impedance microphone or guitar pickup signal, while the audio coming out of a digital audio player should be fine driving as low as 20Ω.
5k is probably a safe bet, but I'm a noob at proper audio circuits.

>an audio opamp with 22k input impedance
Have you considered using a non-inverting topology with effectively infinite input impedance?

>>2517073
The circuit is already built. It uses the two halves of a NE5532 with one non-inverting stage driving the other. Both are set at 22k impedance. I want to add a volume knob in between, it'll be easy I just need to pick the value.
The input has a 100nf dc blocking capacitor, but the output from the other stage doesn't have one. So the knob will come straight from the output, then through the cap. Will that matter?

>>2517074
The potentiometer won't change the cutoff frequency, I even simulated it. 100Hz cutoff is a bit shit for audio though. Less of a big deal if it's for an instrument, I guess.

>>2517076
thanks. it's for an instrument. If I plug my high z guitar with 1 meg pots into a low z input impedance it sounds like it rolls off the high frequencies. I was thinking that a 100k pot into a 10k or 20k input would do the same thing but I guess not.

Hello /ohm/ would you help me out please?
I want to build a voltage regulator like pic related. https://www.bristolwatch.com/ele/lm317.htm
Unfortunately I don't have the 2n3055, I've got a whole bag of MJ3001 TO-3 transistors though. Can I substitute using this tranny? I have no clue which parameters would be of importance here.

>>2517031
>I don't understand

it's very similar to this idea >>2517021 except the cap is in series rather than parallel.
initially cap is shorted so it's not charged.
you apply power, the relay flips, which cuts off direct power to relay.
but the cap is there and (being uncharged) it acts like a short circuit, so it continues powering the relay for a half second until the charging current drops.
the relay clicks off and then the whole thing repeats.

>>2517105
>Can I substitute using this tranny? I have no clue which parameters would be of importance here.

should be fine.
current rating is only 5A instead of 15 but shouldnt make a diff in your situation.
it's a darlington, which is an important difference, but that shouldnt matter either as the circuit is self-adjusting.
either one will likely need heat-sinking.

how efficient are stepper motors at electricity making?
i noticed when i move my 3d printer by hand, leds on the pcb light up
so if i for example connected a stepper to my bicycle could i make free electricity efficiently?

>>2517127
Alright. Thanks.

>>2517130
it would work but stepper motors as generators aren't efficient relative to conventional solutions afaik. also it's "free electricity" only if you ignore all of the other inputs to the system. jewgle suggests 100 watt hours for a bicycle generator is reasonable. how much food do you need to eat to sustain that output, and what is the cost of that food? idk what country you're in but, in the USA i can buy electricity for about $0.15 per kwh (or $0.015 per 100 wh.)

>>2517070
it's plugged to a BMS

I need to control about 20+ led strips independently and simultaneously. Im going to use an esp32 for bluetooth and it has 16 pwm channels. I might be able to pull of using its 8 rmt channels but before I go down that route I wanted to know if there was a simple IC that lets you send multiple pwm signals at the same time.

>>2517275
I know nothing about electronics, but I think you could use an FPGA to like draw a matrix of LED's.
I am only saying this so that you could probably just copy some one's elses work and modify it to work for your problem.

>>2517275
RP2040 can do 16 PWM channels simultaneously and independently. Pico W integrates BT & WiFi.

>>2517274
Yeah but does the BMS actually regulate current? Most I've seen have shutoff limits and such but aren't designed for actual charging. Even if it does chances are it will be linear regulation instead of having an in-built SMPS, which means it will heat up and not have much current capability.

>>2517312
the bms only does voltage protection during charging, i connect it with the switching supply (40v 10a) which is regulated.
40v is used to charge 36v, and I assume that for every battery in parallel will divide the current (I had a feeling 10a is too much anyways, because 10a is the rated max current a single cell can charge with).

>>2517312
>>2517326
Oh, so batteries must have a constant current to recharge... That's important to know.

>>2516908
>I've only ever seen one thing in person that uses LiFePO4s, which was a wifi doorbell for some reason.
they needed the operating temperature range of LiFePO4

>>2516100
Ok, mystery solved, putty is just sending 0x0d, 0x0a, carriage return, new line, no idea how to get it to not send that when you press enter, but I've changed the program to expect it as a possibility.

>>2517407
Check your "Terminal" settings in putty and uncheck "Implicit CR in every LF". Check "Auto wrap mode initially on". In the "Session" tab, select the "Raw" protocol in the dropdown box, then click the serial button and connect.

>>2517065
ok maybe I will use voltage regulated 36v power supply and wire it to a DC to DC buck boost constant current module to bring it up to 40v like this:
https://www.aliexpress.com/item/1005004334186734.html

>>2517435
Just current regulation alone is insufficient for safe lithium ion charging. You usually want some endpoint that cuts off once the CV-stage’s current gets low enough. Do some research on proper lithium ion charging profiles to see if that’s important. It’s possible that the BMS is such that it will cut off automatically, though I don’t know if I’d trust that.

what is the cheapest way I can log voltage to a graph and view it.
it needs to log potentially hours of voltage

>>2517531
Knockoff arduino plugged into an existing computer outputting analogue values via serial. Knockoff arduino spitting analogue values onto a text file on an SD-card would be the cheapest if you can’t waste a computer. Arduino costs a couple of dollars, SD card “module” costs a dollar, the card itself costs 5-10 depending on how sketchy you’re willing to go. Can make the microcontroller cheaper by going for a raw micro, and a flash memory IC may be cheaper and big enough, but that increases hassle by quite a lot.

>>2517536
ok whats the least effort way for under $50
also I should clarify I need at least 1khz read

>>2517531
unless if you are dealing with low voltages you can do what anon above mentioned with an arduino.
for higher voltages, there are multi meters that have software to connect to your PC, but generally they are for very expensive meters, and sometimes need to buy the adapter separately (and depending on the era the multimeter is made, and if there is any open source projects, you may need to use a VM with an old version of windows to use the software).
Downside is that some multi meters have a pretty slow polling rate.
I wouldn't be surprised if there were oscilloscopes with a similar feature, not sure if the ultra cheap digital oscilloscopes have a PC interface, let alone a logging feature.

>>2517545
wait i have an analog oscilloscope
can i set it so that if the voltage drops below a value it freezes and shows me that drop +- a second or something

>>2517550
I think you need a digital one (some analog ones are both analog and digital) with a megabytes rating for how much it can store when your trigger is activated.

>>2517542
Yeah arduino + SD card module is least effort and like $15. 10 bit accuracy built-in, you can get a 16-bit ADC module for another dollar if you need more accuracy.

>>2517545
For higher voltages just use a voltage divider. Maybe optocouplers between the ADC and the arduino if touching the card could give you a shock. Another dollar module. Lower voltages are more of a pain since you need an amplifier.

>>2517550
Analog scopes have no storage feature. Even if you have a hybrid or digital scope it’s a bit of a shot in the dark.

>>2517557
>>2517536
ok uh are there any guides i can follow for the adruino based one then?
also its 12v, the resolution could be 0.01. I just need to know what this 12v bus is doing at the time of a failure, if it is drooping or cutting out momentarilly. Software isnt catching anything.
it needs to be enough polling rate though to catch any spike large enough to cause problems which I have no idea what that is

>>2517581
To measure 12V with a 5V ADC, you need to step it down with a voltage divider. This is just two resistors, 15kΩand 10kΩ would do the trick, giving 4.5V maximum when given a 12V signal. Google image search "voltage divider" or "adc voltage divider".

0.01V LSB at 12V is a tad too much range (1200) for a 10-bit ADC (1024 max). But a 12-bit or higher would definitely work. The ubiquitous ADS1115 is a 16-bit ADC for $2, but has a maximum sample rate of 860 samples per second, so probably isn't good enough. The also reasonably common ADS1015 is a 12-bit ADC that can handle 3300 samples per second. So I'd go for that, just make sure you get a 1015 and not a 1115.

Also the SD card module may need a capacitor on its power pins, I had trouble with that before once. 10µF with at least 6.3V rating is what I'd go for.

So your bill of materials is:
>arduino (nano probably)
>ADS1015 module
>SD card module
>some resistors and capacitors

As for the code side, there will be tutorial code segments on the arduino IDE for both those peripherals. I know there's one for the SD card side about datalogging, just copypaste bits from ADS1015 tutorial code to replace the analogRead() stuff. Delete the serial bits from the code in order to run it away from a computer.

Wiring should be explained in the example code.

is there any way to modify an ordinary wall light switch so when it's pressed it generates enough power so i can power small micro for a second or two and send out wireless data packet over lora or something like that?
the micro only consumes 1ma and needs about 2ma, the lora module is more thirsty i think it's like 150ma at 3v for about a second, so i would probably have to use something less thirsty.
But the main question is, what sort of power is it possible to generate from a single wall switch press? how many watt hours? is it in the ballpark for what i need to do at all?

>>2517665
>needs about 2ma,
2V

>>2517665
Watch this:
https://youtu.be/9Pw7U0XFgUM
This one uses a magnetic transducer, but another of his videos about a kinetic switch uses a piezoelectric transducer. The magnetic ones have the advantage of better being able to tune the output voltage (changing the windings of the coil), but are more mechanically complicated.

You may be able to buy a kinetic switch that looks like and has the same mounting holes as a normal light switch. These likely won't do LoRa, but modifying it to do so wouldn't be too hard, the mechanical side is the hardest part.

good AliExpress store to buy a few non-explosive/incendiary Li-Ion 18650 cells?

https://www.aliexpress.com/item/1005001818722702.html

https://www.aliexpress.com/item/1005002059492402.html

https://www.aliexpress.com/item/1005001984441459.html

Anyone have experience with these kind of units? Are they basically a buck/boost module with an LCD and a rotary encoder?

Is there a better alternative for the same task/form factor/price?

>>2517665
>enough power so i can power small micro for a second or two

there's a smarter way to do this.
timer switches and thermostats are wired in series with the load, and sip a little bit of current thru the loop to power themselves.
i havent looked but i imagine when they're on they have a small resistor in series to generate a voltage.
when they're off, they have some diodes and a high-value resistor to do the same.
using 5 diodes and 2 largish caps you can get 1.95V to power one thing, and 3.25V to power the other.

>>2517716
there won't be any power wires going to the switch so that is not an option

>>2517718
Why do you want to do bad things, Anon?

Is there a way to look how WiFi code works on ESP32 internally? I assume it's running as a task in FreeRTOS? Is it possible to inject my code inbetween?

I have a Wii Rock Band keyboard that I want to use with a PS3. I have read online that all you need to do is remove a single resistor to make it a PS3 keyboard, but would it follow that I would need to remove the same labeled resistor on the dongle itself? I haven't been able to find any confirmation online, though I have found a similar post where someone converted a drum set between consoles, and they modified the same resistor on both the instrument and the dongle.

>>2513810
Would something like OP's picture work with regular AA and D batteries?

>>2517835

a D cell can put out 5A for a few mins, so with 12 cells (2-parallel, 6-series) you could play ''god of fire'' for a lil while.

Don't know if you remember me, but I'm still at my shitty DIY amplifier.
I can't into symetrical power supply, so I'm using a regular single transformer that outputs 19v, recitified thats around 26v.
For powering the amplifier I built a linear power supply regulating down to 24v with around 4000uF stabilization.
I still want to use that TDA1521 that I found in some wastebin since something like a LM3886 is way overboard since I'd need a bigger transformer and it's all to much for my first DIY amp, I think.

Now studying the datasheet of my chip more closely raises some questions. Does pic related mean that 12w is only possible with 10% THD?
Would this be very noticable to normal ears?
I suppose that chip is quite outdated by todays standards, do you know of any similar single chip stereo amp in the sub 30W category?

>>2517127
Supply works fine with MJ3001 btw :)

>>2517858
You should be able to get about 10W clean from a 24V supply. Theoretical maximum would be like 18W into 4 ohm.
The cheap chinese lm1875 boards are actually good, just make sure you get some legit ICs somewhere else.

>>2517861
>Theoretical maximum would be like 18W into 4 ohm.
That would be the per channel, with 10% distortion, right? My speakers are 6 ohms.
>lm1875
They are quite expensive here, will overthink it.
General question: Since they are quite flexible in voltage supply, would I still need voltage regulation, to make up for changes in the mains voltage?

>>2517868
Just checked digikey, about 5€ per piece. That's alright, I guess.
https://www.digikey.de/de/products/detail/texas-instruments/LM1875T-LF02/3640530

>>2517868
24V p-to-p = 8.48V RMS. that's the theoretical max for your power supply
you probably will not need a regulator if your transformer is decent

>>2517810
what do you mean? how is a wireless switch bad?

>>2517880
Sorry, I was projecting. Ask Jeeves about "Zenith Space Command", then take your vitamins and say your prayers.

>>2517861
Wnat's the difference between lb05 and lb02?
Should I use two seperate power supplies to create a stereo configuration out of them? I don't want to use premade PCBs I have access to a CNC to make my own PCBs and want to include some nitpicks like some sort of EQ.

I want to use an nchannel mosfet to control 12v with 3.3v. I read that FQP30N06L and IRLB8721 were similar so I ended up buying the IRLB8721 but now Im reading that it doesnt work with 3.3v. So before I buy the FQP30N06L I wanted to ask if that one would work with 3.3v

>>2517904
familiarize yourself with pic related. in this case, the lowest Vgs they show is 3v, with about 10 amps at 12 volts. However, the other one says it's good for about 4 amps at 12 volts, driven with 3 volts Vgs.

>>2517910
chart for IRLB8721

>>2517910
>>2517911
And of course the Vds in your case will be whatever is left over after the current passes through your load.

>>2517921
i'm currently passing my load

>>2517065
When a power supply says that it's gonna make a audible noise, how loud is that going to be?
https://www.mornsun-power.com/html/pdf/LM200-10B48.html

https://www.noritake-elec.com/technology/general-technical-information/vfd-operation
good website

I just ordered a Pinecil, but I realized it doesn't come with a power supply.

Did I get memed? Which one should I get?

>>2517980
you can look at the used market and buy a laptop charger, or you may even already have one.
if you have a old car battery, you could use a cheap step up converter to use it.

>>2517911
I see, 2A should be enough

>>2517980
Pinecil V2?

>>2517980
imagine not having a 67w USB c charger
but really you should have actually done even a basic amount of research on the product you bought for $40

hello, anon with the simple led circuit again
I had an exam so I didn't reply

>>2517127
are the assumptions in picrel correct then?

in 1. current goes through inductor in relay, relay switches to other port

in 2. capacitor acts as short circuit, current goes to LED (lighting it?) and inductor until Vc = Vs

in 3. because Vc = Vs, current no longer goes through capacitor, which stops feeding current to inductor, so relay switches to original port

and sort of 4. but I put it in 3., a voltage drop is now possible and so the capacitor discharges to the LED

am I correct in assuming the LED is thus lit twice per cycle?

>>2517900
>What's the difference between lb05 and lb02?
Sounds like different batches or something, IDK, read the datasheet. Could also be a different footprint so be careful.
You can also look at TDA2030 chips, but be aware there are a lot of fakes.
You don't need to go with separate supplies unless you want to. There's lots of options.
The most critical parts are going to be the power transformer and the heatsink, anyway.

>>2518042
>are the assumptions in picrel correct then?

1 yes
2 yes
3 no. Vc never becomes equal to Vs (whatever those are)
what happens instead is the current thru the cap falls below the ''holding current'' of the relay, so it flips off and LED goes off.
4 no. the cap is shorted by the relay when it clicks off, resetting everything to initial conditions.

>>2517604
I'm another anon. Can you make some sort of oscilloscope with this ? Let's say I have a pico lying around, can I get precise measures of the shape of =<5V signals ?

>>2518146

https://how2electronics.com/how-to-use-adc-in-raspberry-pi-pico-adc-example-code/

>>2518146
Yes, and people have done that sort of thing with a range of different microcontrollers. Ultimately, there's a few things that will be required to bring it up to par with a proper oscilloscope.

Firstly, sample rate. The maximum frequency you can properly measure is equal to half the sample frequency, so with the RP2040's peak sample rate of 500kS/s you can get at best 250kHz. Which is definitely better than some MCU-based scopes I've seen, but far from the ~50MHz you'd expect from a purpose-built scope. Increasing this is possible, but means implementing a high-speed ADC and the analogue front-end to get it working at full potential.

Secondly, voltage ranges. The voltage resolution is a function of gain/attenuation, and of the bits of the DAC. To get any kind of decent range (say, from 40V full-scale to 16mV full scale like my scope), you'll want multiple gain/attenuation settings. So mechanical (or analogue) switches and a handful of precision passives and operational-amplifier ICs to ensure linearity.
As for the bit resolution of the ADC, I'd personally want at least 10 bits so the quantisation error is less than 1/1000. 12 or more bits would be a nice premium. If you'll be using it for audio, 16bit is likely what you'd want.

Thirdly, data rate. Whether you're plugging it into USB, logging the data into a memory chip, or displaying the traces on a connected display, you'll need the hardware to shuttle sufficient data. The RP2040 can likely function as a scope shuttling ~8bit data from its ADC to any of those without much issue. If you want faster sampling however, you may well run into bottlenecks. The USB controller can only handle at most 12Mb/s, and display driving might well be slower.

Finally, input protection. Especially for a USB scope. Blowing up your hardware because you swapped some wires around is a bad look. TVS diodes, optocouplers, relays, crowbars, etc.

>>2518151
>>2518186
Thanks, mates. I've been wondering. How do you find the components that meet your requirements ? I'm just aimlessly looking through supplier catalogues, and it usually proves fruitless.

first stripboard design
it does require cutting between holes, we'll see how that turns out

>>2518187
search for parts that have plenty in stock and aren't too expensive. get reading datasheets, should get an idea of the price/performance curve. then refine your specifications knowing what you actually want, and what you think is worth aiming for. repeat until done or you take so long that all the good parts are out of stock.

>>2518060
First time ordering from Digikey. Apparently if I pay ~20$ more, I get free shipping - the same as I would pay for shipping right now since they ship from the US apparently.
Any recommendations for stuff to cons***************m? What are some cool things to build/ useful things to hoard for the next current thing(tm) chipshortage or whatever?

>>2518234
>The latest PicKit
>WS2812 LEDs
>crystals
>a selection of modern power MOSFETs
>good mosfet half-bridge drivers
>inductors
>lithium battery charge control and BMS ICs
>switching controllers (TL494s lmao)
>a bunch of audio connectors
>a selection of nice BJTs for power amps
>some good audio-quality op-amps
>audio-quality I2S ADC and DAC ICs
>nice comparators
>TL431s
>radio demodulation ICs
>IR receivers
>74LVC1Gxx single logic gates
>small-signal MOSFETs
>ceramic capacitors you can trust the dielectric of
>CD4046s
>ICL8038
>BBDs
>JFETs
>toslink sockets
>nice slide-pots
>quality solder and flux

As for project ideas:
>sequencer
>drum pad synth
>a bunch of amplifier topologies
>function generator
>CC/CV adjustable power supply
>stroboscope
>LCR meter
>milliohm meter
>quality USB microphone

>>2518234
I use mouser and tayda, sorry. You're already paying shortage prices for chips. Do you have power resistors to test your amp? Don't skimp on the layout or the extra components, the chipamp chips tend to oscillate if you do. Make sure you get the grounds in order.
The top one in your cart has the alternate footprint.

Does this mean I need an asymmetrical supply for that?
I just want to build a simple test circuit to test the fake lm3886 I got.

>>2518251
*symmetrical

>>2518248
>Don't skimp on the layout or the extra components
I live in a socialist country and am doing my EE cert in a facility for subhumans - mostly retards and autists - where I can do whatever I want most of the day and have a CNC mill and a big workshop with all kinds of components readily available. So getting decent components shouldn't be an issue.
Downside of all that is that I don't know what to build or occupy my time with most of the time since there are so many options available. That's why I'm shitting up this thread with half assed retarded posts. Life is pain.

>>2518257
Premise is that they can literally get every retard who has no passion for electronics a certificate in EE. It is depressing.

Any chip can be used on a single supply but you have to bias to 1/2 and block the DC. Read the datasheet there's a schematic for it in there.

>>2518251
>Does this mean I need an asymmetrical supply

you seem to have a severe hangup about dual-polarity supplies.
they're not icky like girls, or terrifying like clowns.
it's just 2 diodes, 2 caps.
fucking pussy.

>>2518340
I meant symmetrical power supply, but what you posted looks like what I'm looking for. I dunno.
I'm propably retarded, I just coun't find any coherent information on that on the internets.

I also asked my master craftsman about it and he told me I'd need some kind of special transformer or two identical tranformers in series of which the middle of the windings wuold be 0? I dunno lol.

Hello, I got an issue with gerber PCB milling, not having any experience with that myself
SO I got these gerber files, but the CNC mill somehow doesn't want to mill it. Coworkers said there was some sort of issue with the layers not being clearly defined. There can obviously only be two layers for it to work with the mill. Any idea?
As I said I have absolutely no clue how this works. Can I fix those files?
Can you please tell me what's wrong with them, Anon?
https://files.catbox.moe/4tc9b5.zip

>>2518353
How is it in 1998?

>>2518340
>>2518348
So dual polarity PSU and symmetrical PSU is something different, I suppose.

>>2518355
Pretty comfy desu

>>2518356
>So dual polarity PSU and symmetrical PSU is something different,

dual polarity = both positive and negative outputs
symmetrical = the positive and negative voltages track each other, or are same value
so the drawing above is both

>>2518348
>I'd need some kind of special transformer or two identical tranformers in series

that's called a center tap transformer: 3 wires on the secondary instead of 2
it's the most common way to make bipolar supplies, but it's not the only way.

>>2518362
>voltages track each other
I still don't quite get it. Gottverdomme. What do those 500pF(?) capacitors do?
>>2518367
Alright. So a center tap transformer would be the better solution, but I could do without one using those two diodes, but I'd need a lot of stabilization then ...I suppose.

I still don't completely get the reasoning behind it, though. I guess it's something about different parts of the amplifier circuitry being powered isolated from each other?
But I thank you for clearing some things up. Unfortunately it's still quite a mystery to me how all that works. AC in general is still a mystery to me, so I guess it's better to not bog you any further with my questioning.
Do you know of some more detailed explainations elaborating on the issue of dual/symmetric power supplies and all that stuff?

For now it's good enough that I know how to build up such a symmetrical supply and I'll try powering that chip amp using this configuration.

>>2518353
sounds like you didn't think that through at all

>>2518374
>What do those 500pF(?) capacitors do?

the voltage coming out of the diodes (whether it's 2 or 4 diodes) is a rectified sine wave.
to make it DC, you use large caps to smooth out the big bumps and just leave a tiny bit of ripple on top.
the min size of cap needed can be approximated as 2000uF for every amp of current used by the load.

>>2518376
Yes, but when it's only 2 diodes it's only halfway rectified, right?

>>2518375
Well, I just gave them the PCB zip I downloaded off the Hackaday project page. I though there would be all nesseacary files in there.

>>2518246
Thanks for making that list. I'll think of something.

>>2518377
>>2518376
Alright, so the caps are the smoothing caps then. One for the 'negetave' and the other for the 'positive' output.

>>2518379
normal pcbs aren't just milled, they're also etched
you're gonna have to convert the files somehow

>>2518383
I thought a gerber file is all that's needed. So it needs special gerber files giving instructions for the mill to operate?

>>2518377
>hen it's only 2 diodes it's only halfway rectified,

yes, and it means you should use caps that are twice the size of the 4-diode version.
the advantage of the 2-diode system is that the transformer is very common, so you prob have one already.

>>2518386
>and it means you should use caps that are twice the size of the 4-diode version.
Right, that was my guess in >>2518374
Well, I'll go with something around 6000uF then, maybe even more. I have loads of big ass caps in the workshop.

>>2518386
Is the polarity correct?

>>2518392

if you want it to work, it's correct.
if you want a small explosions, then no.

>>2518392

>>2518398
>>2518401
this thread is comfy
i like you guys

>>2513810
Question: I have a laptop and PC. A keyboard is connected to the PC. A random usb wire is laying near the keyboard. When I touch one end of the wire to the laptop's USB port the USB keyboard that is connected to the PC disconnects and reconnects. Whats happening and how to fix?

>>2518445
Random usb cable* - its not connected to anything just USB to USB-C cable for a different keyboard.

>>2518386
Oh dear. the downside is that the resulting voltage will be basically unusable with the transformer I got.

>>2518461
A: avarage is useless compared to RMS
B: that doesn't matter since you'll be filtering it with big caps
C: it will be twice that anyhow because you've got two rectifiers, one outputting +(Vpeak - Vdiode), the other outputting -(Vpeak - Vdiode)

>>2518463
Alright. I'll try it out in the lab soon anyways.
I just want to finish this shitty amplifier already and using the TDA1521 I want to use that wastebin transformer that I have no other use for. Basically a scrap amplifier.
Guess I'll build something more high end using an lm1875 or descrete components later on.

>>2518461

looks ok to me.

>>2518468
Yeah. That calculator is bullshit.
Oh and we have 50hz in europe. I don't know if that would change much about the rectified output.

>>2518468
>more than ±18V
that could actually be a problem

>>2518484
Schottky diode voltage clamp

>>2518485
You mean a zener? A pass regulator would be better for thermal reasons.

>>2518486
>You mean a zener?
Yes. My brain is melting.

>>2518484
>a problem

TDA1521 datasheet says it's good to +-21V

>>2518490
Never mind then.

Tonight I'm going to learn how digital radio works.

>>2518500
CMOS PLL?

I'm learning about recaping and repair.
I took some old VIA EPIA-CL mini-itx out of my broken devices' crate, tested caps with ESR. 24 out 36 are dead so I removed all caps and consigned refs in a table.
Where should I dig for good quality caps ?
Can you recommend me a good electronic pump for tin ? Working with hand pump on mini-itx reveals fast its limit.
If, by instance, after recap there is no return to working state, what are other potential parts to dig & replace on a mobo ?
Last question maybe, where could I find a parts' schematic for that board ? User Manual don't provide any.

Last question, I'm going then to recap some 8bits and 16bits. Are ebay vendors selling good quality or it's just random caps ?

>>2518513
Mobos are tough to repair exactly because of how complex they are with a low chance of finding a schematic. They're not really designed to be serviceable, even by the manufacturer, especially for newer boards. And if it's an older board, it probably isn't even worth repairing. All those caps would likely cost more than a 2nd hand board on eBay with the same specs. For quality caps, you likely want to buy proper brands, new, directly from reputable vendors like DigiKey, RS, LCSC, Arrow, that sort of place. Alibay vendors can sell good caps on occasion, but the risks are too high and the value per dollar is almost certainly going to be worse than a proper order on Digi-Key, if just because all your shipping will be collated in one lot.

I made a fool of myself tonight as I spent the whole evening trying to solder two welding rods (stainless steel) together for a Christmas present. I followed the wikihow really well, had strong flux and scrubbed the surfaces with alcohol. Tested both lead and lead free solder but will find another solution tomorrow.
Not really /ohm/ but have you successfully soldered stainless with an electronics soldering iron?

>>2518579
Did you sand them through olive oil? Apparently that works for aluminium, I expect it to work for stainless.

>>2518581
>sand them through olive oil
not falling for that one
while taking a dump just now I remembered having some scrap solid core cable I will try instead,

Just now realized I have seen it done with marble tracks and behold,
https://youtu.be/kPguktA674w
using sandpaper instead of the steel ribbon scrubber I tested must be the key to success!

Sorry for the doubt, but I'll try without the olive oil first tomorrow. Of course the steel scrubber wouldn't bite on the filler rods!

>>2518594
So long as it's covered in something to prevent oxidation while you're sanding it, there's basically no reason for it to not work. Just sanding it dry and applying some rosin paste afterwards wouldn't work for aluminium, but it might be ok for stainless. Putting rosin paste on your sandpaper might be easier to clean off the sandpaper without destroying it.

Remember, electrical solder isn't very structurally strong. Also you need good enough thermal contact to heat the metal up to the right temperature.

Hello anons.

I need a circuit that can pickup the pulse of a magnetic pickup(about 1v AC pulses, RPM dependent) and turn it into a 5v square wave. It's going to be in a noisy environment(under the hood of a car). I plan on running everything inside an aluminum enclosure but I assume I'll also need some circuitry to filter out noise too on the line leading to the pickup itself.

variation in trigger time does not matter, so long as the variations are consistent with RPM.

>>2518708
Limit

>>2518708
Where is the AC signal with regards to the voltage rails? So long as the signal is within the common-mode range of a comparator (consider a RRIO comparator), I'd use a schmitt trigger comparator, pic related. The first trimpot is for adjusting DC bias, the second is for adjusting hysteresis. Should help with noise immunity. You may also want some filtration, be it for eliminating noise, or for normalising the frequency/voltage response curve of the magnetic pickup (usually higher speed = higher voltage). An integrating op-amp circuit may be necessary instead of a passive filter if gain is necessary at low frequencies. A high-pass filter may be necessary if your signal is about the 0V rail and you don't have a negative rail.

>>2518708
I'd just use some ESP32 and PWM this bitch after some DSP

>>2518484
It could deliver almost 16w at 8 Ohms. That's just what I'm looking for.
The datasheet doesn't give any charts for other impedances, but using 6 Ohms, could it use the full 20w ? I suppose that would pretty much exceed what my 50va transformer can deliver already.

>>2518725
blinkie moment

>>2518771
Probably not. I'd simulate a class B amp and see what power dissipation it has as a function of signal amplitude, source voltage, and load impedance. I tried doing it analytically but found it was a pain in the ass. Higher voltage rails (or lower speaker impedance) means higher voltage dropped across the amp IC for the same drive current, and hence increases the power dissipated by the amp. But my equation didn't show that at all, idk. Neither does that graph.

>>2518539
I understand. Thank you.

And, excuse my noobiness going from scratch, considering all those SMD capacitors of extreme small size, are they as well prone to become "leaky" ?
I just tested a few of them near the PCI port of the motherboard and received also "leaky" answer from the ESR.

>>2518779
SMD electrolytics are no different to THT electrolytics, aside from the shorter leads and hence lower ESL. Same for tantalums, though tantalum SMDs often come in somewhat larger values. SMD ceramics come in much higher values since they have multi-layer construction, but also can have issues with voltage-dependant capacitance, microphonic response, and fractures from board flex. The former two are a function of dielectric choice. Ceramics have lower ESR and lower leakage than electrolytic capacitors.

What kind of measurement did you take? Was it measuring series or parallel resistance? If it was in-circuit it could have been surrounding components, though testing out of circuit can be difficult. If they're ceramic caps, I'd usually assume they're fine. Finding their values to sort out a replacement is tough since they usually don't have values written on them anyhow. If they're tantalum caps, they can go bad over time, electrolytics even more so.

>>2518773
I will when I'm in the workshop again this week.
What is some good simulation software for ganoo / lunix?

Also, I came across this term "gainclone" recently, that is apparrently used for DIY chipamps.
Does this term only include TI overture chips? Because Wikipedia also isn't clear on the issue:
>Now, the term "Gainclone" could apply to almost any amplifier based on a power integrated circuit of some type, from any manufacturer.
Now what's so special about this "Gainclone"? That single chip amps prior to the 3875 don't hold up to audiofag standards?

>old iron dies, get a new one
>comes with a good tip, durability is great but the thermal transfer is a bit shit so it's a struggle to desolder stuff
>buy some new tips
>really good thermal conductivity, any joint melts in seconds
>30 minutes of work and the tip already has big craters in it
The duality of tips

Grabbed two old scopes for free when my uni was clearing out old equipment. I got the second one for my friend.
A quick google search shows these selling for around $150-200.
Looks like Christmas came early anons!

Anybody here have experience with these scopes?
Any chance I could use the printer port on the back for printing to pdf or an image file?

>>2518783
>What kind of measurement did you take?
simple ESR70 gold with 2 needles on each sides of some tiny SMD capacitors.

Very enlightening answer btw, thank you.

>>2518374
I also found you can make a regulated symmetrical power supply using an OP Amp.
But I guess the TDA1521 doesn't need that regulation. I'll add this functionality to the high current regulator I built earlier anyways. >>2517105
http://www.dl8stwblog.freiraumwelle.de/?p=2170

why isn't eevblog in the paste?

>>2518829
>Anybody here have experience with these scopes?
yes
>Any chance I could use the printer port on the back for printing to pdf or an image file?
you're an idiot.

>>2518829
where do you live? I'd consider buying one if you're local

Anyone ever ordered components from win-source.net?
I need to buy some obsolete ICs (Altera EPM7032SLC44-7) and they're the only supplier I can find that claims to have them in stock, but I never even heard about them.

>>2518834
>you can make a regulated symmetrical power supply using an OP Amp.

just no!
that's only good for low-current situations, like in a guitar pedal.
if you're driving several amps into the ground, an op-amp is not gonna take that.

>>2518879
>deleted post
Warrrumäääh?
What did he mean by that? Is an OP amp good for high cuurent situations now?
Can't you somehow use a high power transistor in combination with the OP amp?

>>2518906
https://wiki.analog.com/university/courses/electronics/text/chapter-4

>>2518906
>Is an OP amp good for high cuurent situations now?
the op-amp specified in the drawing is especially made for that purpose, it's good for 3A.

I'm a retard. How do I get something like a relay (switching independent loads on/off) but with 3.3V (MCU logic levels) on the control ports? Most actual relays want 5V. Needing a separate 5V power source isn't nice in the first place. It doesn't have to be a relay. Use a solid state relay? A MOSFET? Is there anything else?

i need a simple signal generator for testing shit like audio circuits
i found this, seems pretty good
https://www.aliexpress.com/item/32862689682.html
Before i buy, Is there anything better, but also cheap that chinks sell?

>>2518915
Take a look at this piece of shit website lmao
https://www.circuiteasy.com/what-is-a-relay-driver

or look up "relay driver"

>>2518921
>i found this, seems pretty good

wonder where they found an XR2206 chip, which was discontinued in 1991.
according to some random guy on the eevblog forums, they're fake and they're defective.
anyhoo, if all you really need it is for audio, you can generate waveforms with audacity.
if there's no computer near you, make recordings at various frequencies into an MP3 player or phone.

>>2518915
>Use a solid state relay?

trigger voltages can be as low as 3V, so find one that specifies that.

>>2518633
Thanks, I made a couple tests again using the abrasive side of a sponge instead of the stainless scrubber from yesterday but to no success, see top left. It became a little better after scrubbing with olive oil (bottom left) but popped off really easy.
I'll have to use the coiled enameled wire. Some imperfections (as it's not perfectly straight) will add that SOVL for a homemade gift I guess.

>>2518953
Clamp them together, hang it from a piece of wire on an s-hook and torch it.

>>2518717
Thank you for the response anon, but I have a few questions because I'm retarded and relatively new to electronics.

First off:
>Where is the AC signal with regards to the voltage rails?
What exactly do you mean by this? The AC signal is generated on it's own with no relation at all to the voltage rails. It doesn't even share a common ground.

Also, I take it that the resistor symbols with arrows through them are trim pots, but what exactly are the two top triangles and what the fuck do I even do with them?

for further details, the magnetic pickup is inside a distributor, which will be ...distributing... ~25-30,000 volt pulses to fire the plugs. I want to read the mag pickup sensor in order to get a reading on RPM.

>>2518953
>enameled
*galvanized

>>2518959
Not investing more than time from now on otherwise a welder or a powerful drill to straighten the galvanized wire would be nice!

>>2518968
>galvanized copper
huh? Also, NEVER heat galvanized metals. It will fuck you up permanently.

>>2518934
>they're fake and they're defective.
typical
i guess i could just take atmega328p or esp32 i have and connect digital pin to op amp and then basically fake sines and and stuff by just switching the digital pin on and off really fast

if i have electret microphone capsule what sort of voltage should i expect from it roughly?
no tranny for amplification, just the bare capsule

>>2518794
LTspice should run on linux. There's also falstad.
Never heard of a gainclone, maybe /g/ would know.

>>2518829
If the port on the back is a DB9 or DB25, it may be designed to output RS232 to a plotter. In which case you can just read the HPGL with a USB-to-RS232 cable (might require a null-modem). I did the same, and wrote a little python script to write the data into a .hpgl file. A few alterations were required since it uses some strange special characters (µ, 10x, etc.) that I had to manually write replacement code for the characters they stood for. Once I did that the files were simply openable via hpgl-view, another piece of software written by the dudes over at CERN IIRC.
If it's outputting some other signalling system via serial, then that's a more intensive process of reverse engineering the protocol. If it uses parallel, you may need custom hardware, could be a fun microcontroller project.

Is there a GPIB on the back too? They're apparently real handy.

>>2518834
Yeah op-amp regulated rails are nice and stable, but they usually can't source too much current. So if your speaker is connected between the amplifier output and the ground rail, it won't work. If it could supply that much current, then it's equivalent to your power amplifier IC. You'd need to either have a proper split rail power supply circuit (centre tap or double rectifier), or a second amplifier feeding the other side of the speaker. Which is sometimes done. Or just pass the speaker through a massive capacitor like the single-supply example circuit tells you to do.

>>2518836
Think he's in the comprehensive list of resources.

>>2518934
I found ICL8308s (clones?) on LCSC, bought 2 a year or so ago. They're also kinda hard to come across.

New thread plz
T. Phone poster

>>2518962
If it's not referenced to anything at all, then you can ditch the first trimpot and do something like this. Both Rs are the same value. The arrows now labelled with V+ represent the top voltage rail. So +5V in your case.
As I said, an op-amp integrator would solve the speed dependant voltage I'm expecting you'd get, but you could also replace the pickup with an analogue hall-effect sensor, I think they're fast enough. For that you'd go back to the first circuit with its middle trimpot.

>>2519143
It's page 4, it will be fine for another day or two.
t. OP

>>2519156
Thanks. looking into Schmitt triggers now. I have a really old scope but I haven't hooked the distributor up to a drill to see what sort of voltage peaks it deals out yet. Since this is such a low voltage and it's so close to high tension spark plug leads I should also be looking into filtering as well, yeah? I assume most OEM ignition modules have filter circuits in them to prevent false triggering. too bad 99% of them are potted with a tarry black goop that sometimes melts and makes a mess of the firewall.

>but you could also replace the pickup with an analogue hall-effect sensor
No go. OE ignition module won't work that way. Has to stay mag pickup. If I could have changed sensors I'd have went with a reed switch to avoid this circuit to begin with. Would have been alot simpler for me if I could.

>>2519109
>what sort of voltage should i expect

1 to 10mV.
more if you smash it with your forehead.

>>2519234
The scope would help a lot.

>>2518840
Yes.

>>2518846
I'm in Western Canada but I'm not looking to sell these.
The uni handed out like 20 of these and I overheard a few other students checking the price on ebay and talking about selling them.
So you might notice more popping up on ebay.

>>2519141
The back doesnt seem to have any other ports besides this one. Looking up the model number shows its a "GPIB" interface which ive never used before. If it was an ...8A instead of ...7A it would be rs232. Behind this module is a bigger port.
Googling GPIB to usb comes up with a few results for adapters maybe ill try those.

>>2519326
http://www.questions4steveb.co.uk/html/Other_projects/LPT-Parallel-Port-Oscilloscope

https://web.archive.org/web/20071006063118/http://www.ctv.es/pckits/tutorial.html

If you're getting into soldering for the first time, what do you r͟e͟a͟l͟l͟y͟ need?

>>2519381

roll of kester 44 and a bic lighter.

>>2519381

roll of kester 44 (Sn60Pb40, 0.8mm diameter) and
a bic lighter.

>tfw up all night building Thing only to realize don't have no 9v battery

>>2519492
(frustration)

>>2519493

just pull the one off the smoke detector.
you'll be safe.
i promise.

>>2519381
All you need is love.

fresh from the reflow owen
>>2519612
>>2519612
>>2519612