perfect santa hat fit

Are TL072 prices still retarded or nah

>>2883152
>8c for a SOIC
>38c for a DIP
Seems ok to me, and those are from TI. They’re going to be significantly cheaper still from Chinese manufacturers.

Do guitar pedals need to be in metal enclosures to insulate them and keep interference at bay or is it just for durability
I want to try fucking about with making some but I'd just be making the enclosures out of MDF because it's cheap and I can

>>2883191
Pots with long wires and unshielded casings will definitely act like antenna, which is especially bad for high gain circuits

>>2883191
It’s mainly durability. Check out the grunty foot pedal switches on those things.

> mdf
You can shield it with an altoids tin inside, or some sheet metal for ducts, or a tin can can form a shield.

https://www.youtube.com/watch?v=fs8dVm1Ot3E

>>2883274
explain what we're missing, anon

File: HV PIZZA-1.webm (1.2 MB, 1280x720)

1.2 MB WEBM

Just built a flyback driver and I'm wondering whether food can be cooked using a high voltage arc

>>2883318
tried it once, it just made shit charred on the surface.
the arc is too hot for cooking without mixing its heat with colder air or distributing the heat on a hotplate

might work for caramelizing the top of a creme brulee though

>>2883318
fukk I want a pizza now

File: Screenshot 2024-12-27 110746.png (155 KB, 317x552)

155 KB PNG

is there an upper limit to the number of reference voltages you can have in a voltage divider comparator set up? made a volume unit meter and wondered if i can have more resolution with more IC's resistors and LEDs.

>>2883318
what are you driving it with?

>>2883611
I've got an arduino generating a square wave which then goes through an optoisolator. On the output of the optoisolator I've got a bjt inverter stage to drive the mosfet gate. The mosfet is a 20n50 I had lying around. I've used irf540s before but they would always fail short, probably because I've got a 1n4148 signal diode as a flyback diode instead of a proper snubber.

>>2883610

In practice, you start to run into problems with resistor tolerances (same as with a resistor ladder DAC). In theory, though, there's no limit. You can get quite a few stages by buying low-tolerance resistors and hand-picking the most accurate ones/combining multiples.

>>2883648
> a 1n4148 signal diode as snubber
The diode probably burned out on the leading edge of the first square wave pulse

>>2883138
What is that paste called? And what is he making? Did he order the metal plate stencil thingy or did he custom cut it? What kind of machine can you cut or stamp it with at home? And what is he making with that chip?

>>2883753
1. Solder paste (powdered solder mixed with flux)
1. Volt meters with LCD displays
3. He ordered the stencils with the PCBs
4. a CNC plasma cutter or laser cutter depending on material and thickness
5. Volt meters with LCD displays

>>2883648
nice. ive got a 555 oscillator for when the universe grants me a flyback transformer but for some reason the signal craps out when i connect it to base. surprised your arduino doesn't suffer from feedback issues but that's whats cool about optical isolation i guess. haven't tried 540's for anything but all my 840's failed where the 2n2222 prevailed in my recent slayer exciter. still no break out tho, think my secondary needs more turns.

>>2883648
>50V 0.1A diode for high-voltage power electronics
>putting an anti-flyback diode on a flyback converter
what the fuck are you doing

>>2883769
The 2n2222's low input impedance on the base may be acting to clamp (to an extent) voltage spikes whereas the 840 (or any mosfet) has a very high gate impedance which could be causing voltage to build up and overload the gate insulation. Just my guess as to why the beefy 840 would fail whereas a small signal bjt wouldn't.

>>2883794
>anti-flyback diode on a flyback converter
The leakage inductance of the primary causes ringing which needs to be suppressed or the switching transistor will fail due to excessive voltage. This leakage inductance does not couple to the other winding but does cause ringing which needs to be suppressed. That is my understanding at least, correct me if I'm wrong.

Also the 1n4148 was an afterthought really I couldn't be bothered making calculations for an rc snubber or rcd clamp so I just overspecced the switching mosfet

>>2883802
you're probably onto something, the gate capacitance is probably not draining and then overcharging like you said.. the 840 did oscillate but if i tried to start it with a top load or touch the end of L2 it died instantly.

File: 1734927833926.jpg (1.69 MB, 4000x3000)

1.69 MB JPG

Why is there green shit appearing at my solder joint. It's liquid, cause I can push it around.

>>2883909
That's the dreaded plutonium solder. Aliexpress/amazon?

I am new to KiCad and I have question. I was trying to build a PCB which contains a SOT-23 A19T PMOS. What I couldn't figure out was how to assign the pins to the SOT-23 footprint (the generic SOT-23 footprint have 1=D, 2=G, 3=S, but the A19T PMOS I have had 1=G, 2=S, 3=D on the datasheet) and after dicking around with this for ~30 minutes, I gave up and just created a copy of a PMOS with a new copied SOT-23 footprint and just used that.

Is actually an option to reassign pins from a symbol to a footprint? If so, where? I can't seem to find it anywhere (or maybe I'm just retarded)

>>2883947
Never tried, but can't you remap the pins when you right-click while in the symbol editor and choose the pin table?

>>2883760
Is there a website chinese website that sells a collection of stencils and pcb's like this for many different/common things?

>>2883959
The stencils are meant to speed up low-volume production. If you're just doing a few boards you can apply solder paste without one.
https://youtu.be/xPFujTJbUkI?t=360
https://www.youtube.com/watch?v=ivE6SXDdd0A

>>2883955
I think that's how I ended up doing, but I ran into some issues with saving and KiCad wouldn't allow me to save it, something about a read only copy. So that's why I ended up creating an entire copy of said PMOS. If there is a way to update the pin definitions on the schematic without changing the symbol in the library, then I am just retarded and didn't piece that together

>>2883969
The symbols and packages that come with Kicad are read-only for a good reason. You should not modify them. If you do that, what happens when you update your Kicad installation? Creating a new library for your own components is the way to go in my opinion.

>>2883947
If you don't want to make a new symbol for the A19T FET, you should probably just use the "Q_NMOS_GSD" or "Q_PMOS_GSD" symbol, and choose its footprint later. You could also choose a MOSFET with the same package and pinout, and just rename it.

>>2883974
I mostly agree with this sentiment, but it doesn't stop me from going in and flipping all the PNP and PMOS transistors so the body diode points upwards. Having to flip every damn P transistor just because of some stupid convention is irritating. Should probably make a feature request eh.

File: EL130670_1_en_GB_83ee.jpg (8 KB, 450x450)

8 KB JPG

>>2883928
No but it is chinkshit tier. I've used the cable, heatshrink and solder before and its been fine.

File: Screenshot 2024-12-28 232149.png (17 KB, 663x287)

17 KB PNG

>>2883683
that makes sense, I decided to make make parallel circuits for a spectrum analyzer instead of a long one but i cant seem to get any signal through a high pass filter based on picrel. if i put the cap and resistor both parallel it kind of works but cant see any difference in the outputs when sliding through a tone generator.. it must be doing something to the signal though but really i just need an oscilloscope i guess.

>>2884105
What are your filter's R and C values? What's the input impedance of the signal receptor after the filter, and the output impedance of the tone generator before the filter?

File: IMG_20241229_123713.jpg (427 KB, 1365x1080)

427 KB JPG

Hello. Got this busted kidcamera to mod the fuck out of it. That bloated battery, what is it? 3v7? It has no text. AA for size ref, its pretty small.

>>2884247
Looks like a single Cell LiPo 3.7v nominal

>>2884249
ty

>>2884108
values are 10K and 4.7nf. seems to be a difference in dc resistance about 400Kohm between the amp output and signal input. does this have something to do with impedance matching?

>>2884285
That shouldn't be loading down the signal generator at all, but there's a chance it's giving trouble on the other end. See if you can swap to something closer to 1k/47nF.

DC bias isn't a problem for your signal, is it?

>>2883138
What would be the cheapest way to clean the power from my generator? I need to do some welding away from the shop, and I've heard that inverter welders don't like cheap generator power. I don't really care to purchase either an inverter generator or a transformer welder, since I don't need to weld outside of the shop very often.

colossal assholes, i have a multimeter but no 9v battery so i can't use it. the delivery times are unreasonable because of the holidays. i tried a simple continuity test by trying to short circuit an AA battery, and got a read on the thick wire and even a scorch mark on the connector. now i'm not getting a read on anything even with fresh batteries, the other wires i'm trying to check are thinner so i can't feel any electricity or heat to tell apart from placebo. i don't have loose LEDs or anything to use as an indicator.

https://www.youtube.com/watch?v=J6rEB5_DCZc
kek doesn't look sketchy at all

File: C51C5577-EADD-483F-B0C8-9(...).jpg (732 KB, 786x1078)

732 KB JPG

>>2884406
> broken or can’t figure out how multimeter works
Somehow, you’ve managed to get to the point you’re at now without picrel. It’s more accurate than a DMM and needs no battery for measuring voltage or current.
Be prepared. When the S hits the F, you’ll need it.

>>2884333
You can buy professional “power conditioners”, but they aren’t cheap, and probably wouldn’t solve the problem. My guess is that the old generators vary in frequency much more, and the PFC circuitry inside the inverter welder doesn’t like that. The only thing that can change the frequency of mains power is a VFD, which also isn’t cheap, and I don’t know how smooth the output of one of those is for an inverter welder that’s expecting a sine wave.

I could be wrong though. Oxyacetylene?

>>2884406
>the delivery times are unreasonable because of the holidays
Can’t you just walk down to your local dairy or supermarket and buy a 9V battery in person? They’re frequently used by smoke alarms so they should be widely available, even A23s and LR44s are sold at the supermarkets near me.

>>2884409
check your privilege, it costs at least $2 if my parents will even go to the specific store, the other ones are more expensive. and they just did their shopping today. it's in the middle of the night in my country. and at this point, i could buy a knockoff cable for $12 instead which will be much easier to reverse engineer, i only have the lightning version and i'm trying to make a USB cable. it has 5 wires inside, i can get it to be detected by windows if i connect the 5th wire to +5V but it doesn't work beyond powering on and being detected by name.

and the 5th wire seemingly goes to data (negative) instead of +5V on the lightning connector, but then it doesn't work, i need to continuity test the other wires

>>2884409
I haven't oaw welded since high school, and aren't set up for it anyways, so that's a no. I guess I'll brush up on my stick welding and scrounge up a cheap stick machine off marketplace. Shouldn't be too hard to find an old airco or sears for cheap. Cheaper than the $300/day the local rental place wants for an engine driven unit, anyways.

>>2884411
What are you trying to do with USB, and what is not working? This video may help:
https://youtu.be/VG5bWzEPfsg?si=cvw55NGWQ_-LQwpm

Also get a job.

Currently trying to build this.
https://www.youtube.com/watch?v=-czOV-hKSlA&t=903s

Should I even bother with breadborads anymore? I feel that I spend more time looking around for bad connections that building circuits. Also can someone explain why there is a 100 ohm resistor between the positve rail and the first amplification stage ? I can't figure it out.

File: Screenshot_20241229-235404.png (1.83 MB, 2460x1080)

1.83 MB PNG

>>2884458
this is the circuit I am talking about.

>>2884458
>>2884459
The 100R resistor makes a low-pass filter with that 100uF cap for extra noise rejection, likely from the power supply and the other stages that might be feeding noise into that +5V rail. The current being drawn by that first stage is more or less constant and very low, so the voltage drop across the 100R resistor is minimal. You'll often see this sort of RC power filter on low-current primary amplifier stages.

Though personally I'd just ditch the BJT stages and use an op-amp or two. That transistor amplifier chain looks pretty crude, it all seems eyeballed at first and then tuned until it works. I don't really like DiodeGoneWild's projects. Or his godawful accent. I watched Huygens Optics video on the principle of DD metal detectors. My implementation would be an op-amp receiver circuit with a diode detector feeding a buzzer or piezo oscillator, and a MOSFET ZVS circuit that pumps a lot of power into the soil. It would probably be possible to do phase detection in an audible way so you could tell the metal type by ear, but it might be better to incorporate a microcontroller anyhow for the purpose of sending out short powerful pulses and sampling them, possibly at multiple frequencies.

File: 20241229_223437[1].jpg (2.75 MB, 2992x2992)

2.75 MB JPG

yo, maybe someone here can help me figure out what's not working in this circuit. Pic related is the circuit diagram on the back of the fridge. The plug is fine and so is the source of power (~120v, ~60 Hz) I get 115v on the hot line going to the compressor , I get 155V in the outlet for the light on the inside of the fridge and replaced both the strting relay and the overload protector but still wont start, any suggestion?

>>2884482
Can you measure inside it while it's powered-on? I'd want to measure the voltage from L-C (should be 0V, if not this suggests a faulty thermostat), L-S (should be mains 115V only when motor is starting, dropping significantly lower after), and L-M (should be mains 115V always). Also what does it sound like when powered? Rotating of a compressor, humming of motor windings, or silence? Any clicks from the thermostat?

I don't see a capacitor in your diagram, I'm guessing it's included inside the compressor assembly. If the motor start cap is dead (they do die), it wouldn't start, and wouldn't be replaceable. If it failed short, you'd be able to use an external motor start cap, but I think they usually fail open-circuit. And also shoot their goop out, which may contaminate the oil. Replacement compressors aren't that expensive IIRC, but you may need to get someone with brazing equipment and refrigerant handling capabilities to plumb it in for you if you're not a diytard like HyperspacePirate.

File: 1709122387378651.jpg (282 KB, 2276x1576)

282 KB JPG

This red wire is the factory solder for a battery
I've never undone it, but it may have had some strain.
Does it look like it could pose a problem? I don't know much, but you can sort of see the individual strands(?) of the wire and it doesn't look tinned. Could it be a poor connection or it doesn't matter as long as they're all flowing to the solder point?
The battery does have issues though it's to do with detecting the correct battery level and charging.

Second question, is buying cheap batteries on AliExpress bad? Are they safe or possibly dangerous? It's for a ps controller.

>>2884513
kys retard

>>2884520
rude

>>2884473
Thanks dude.

>>2884458
>>2884459
No. Get a big ground plane and solder everything point-to-point. Individual submodules, like each of those CE amplifiers, can be made by twisting together the legs of their components and soldering, then tacking down the entire unit through either their ground pins or their decoupling capacitors.
Your "breadboard" should be a large copper sheet that you reuse, and once you prove the circuit works you can easily un-tack everything and transfer it to a smaller/cheaper ground plane (copper tape on cardstock is great for small projects).

>>2884596
First time I hear about this. Thanks, I will try it out.

File: file.png (250 KB, 2384x981)

250 KB PNG

i done goofed and put a 431 on a board with the intention of getting +1.3V out of it, forgetting that the 431 can only go down to +2.5V. and i double done goofed because i need to sink 150mA to +1.3V and the SOT-23 431's i got can only do 100mA (i.e. R55 should be going to ground instead of +5V).
options? i can fix the R55 issue with deadbugging and enamel wire, but i dont see an solution to the +1.3V issue.

>>2884631
you're fucked, what's the reference for?
best option i can offer is an op-amp, probably with a bjt output transistor, to buffer a reference or divided reference
maybe you can find a chinky version of the TL431 that can sink more current and go to a lower voltage, but scouring the land for chinky silicon is a bleak task

>>2884503
thanks for the advice, I'm away from the home RN, but will post updates when I get back

File: file.png (18 KB, 304x314)

18 KB PNG

>>2884682
thanks fren <3

>>2883138
short and simple: thinking about building a sound powered telephone. dont fully know how they work, im all for doing the research myself but to save me the trouble whats the difficulty scale on making one of these?

>>2884717
so how do you receive the call
does the person have to shout down the receiver, which activates the piezo whatever in your phone and powers the phone to make it ring?
but then who was phone?

File: xiaomi-microhoo.jpg (226 KB, 2284x720)

226 KB JPG

>>2883138
>Xiaomi Microhoo Air Cooler.
Has anyone ever disassembled this Air Cooler?
How do I remove the shutter knob without breaking it so the external housing can slide up and off?

>>2884765
Knobs are often just friction-fit onto potentiometer/encoder shafts. Sometimes they have grub-screws, but if that's the case you'd likely be able to see it from the side, unless the knob is two-part. Maybe there's a vinyl sticker you can peel off the top.

>>2884766
> grub screw
I haven’t seen a grub screw on a knob since 1967, and that was in old tube equipment from the 1940s.

File: High-Quality-100-Aluminum(...).jpg (106 KB, 1000x1000)

106 KB JPG

>>2884768
They're all over the place when you want to hold something to a shaft. Any half-decent knob meant to slot onto a potentiometer's D-shaft instead of a spline-shaft has one.

File: Knob + shaft.jpg (267 KB, 3050x612)

267 KB JPG

>>2884766
No stickers..
No grub screws.

I was thinking its probably glued..
BUT it'd be better to confirm that before I melt it with a hot air gun at god knows what temperature.

File: inside shaft.gif (1001 KB, 207x134)

1001 KB GIF

>>2884765
>>2884773
video of shaft.

If glue; If I use heat..
What temperature?

I want to attach a set of female pin headers upside down (pins poking up through the unplated side) on perfboard for a pi hat. Wat do?

File: bend over.jpg (20 KB, 400x300)

20 KB JPG

>>2884785
spread those legs

>>2884809
Should work for what I'm doing (only a few pins in a single row) and seems less janky than the idea I had of using hot glue and jumper wires. Thanks.

File: 9976b66516ffad8585a8dc309(...).jpg (215 KB, 1152x2048)

215 KB JPG

etched my first pcb, came out surprisingly well. some of the edges are a bit rough because i used a toothbrush and sponge for developing/etching. i think if i did it the right way by rocking the pcb back and forth, it could be basically perfect for my uses.
>chinky copper clad boards
>chinky negative photoresist
>chinky inkjet transparency
>7.5 watt chinky UV COB lights 4 minutes @ 6" exposure distance
>0.5mm trace width, 0.5mm clearance
>sodium carbonate developer
>fecl3 etchant

>>2884821
Excellent results for your first board. You can use a fish tank with a heater, air bubbler, and UV LEDs in the hood so you don't have to manually agitate the solution.

>>2884829
>UV LEDs in the hood
For exposing the photoresist so you don't have to move from tank to tank and make a mess.

Has anyone here run into an issue where the Wifi on a ESP32 only periodically works? I have this oldish Sparkfun ESP32 Thing V1 and this thing can connect to my Wifi network for like 20 minutes and then just randomly stops. I flashed one of the default example HTTP Server from the Arduino examples and attempted a continuous ping from my main machine and the ESP32 just dies after some random unspecific amount of time. It is strange

>>2884843
I haven't had the issue, but how are you powering it? Have you tried another PSU and/or cable? Is the cable too long? Are you getting a steady 3.3V from the regulator? Are any components getting hot? Any cold solder joints around the USB connector? Debris inside the connector?

>>2884847
This is currently a testing device, so it is being powered from my Desktop. I tried another USB cable, but no success. I have no idea about the regulator, but I'll have to check which part that is once I find the datasheet, but none of the components were getting hot (other than the antenna of course). The only solder joints that are really close to the USB connectors are the pin headers (as I have to solder that on myself), but there shouldnt' be any other debris inside the USB connectors (I used an air duster to dust everything on this thing thinking it was debris)

>>2884851
I'm thinking you have a bad solder joint or two because it works for 20 minutes until heat expansion breaks a connection. You could flux everything and reflow the board, or manually touch up anything sus.

>>2884821
Nice use of copper and preservation of etchant.
Sometimes I turn some of the unused copper into a prototype area and/or heatsink.

>>2884821
Very nicely done. I'm surprised you got such good results out of chinky photoresist, my stuff was covered in imperfections, and my inkjet transparencies really didn't produce enough contrast. I now use a laser to burn away a lacquer etch resist.

You'd be better off using FR4 copperclad instead of FR2 though, it's more resistant to delamination from heat. If you want to do 2-sided boards, you'll want to etch one side, then drill it, then etch the other side so you can ensure the mask is aligned.

What's the project?

File: Capture.png (99 KB, 1887x1190)

99 KB PNG

>>2884883
yeah next time i buy some copper clad i think im going to try to find fr4. can fr2 survive reflowing or hot plate soldering?

im trying to teach myself smps basics so the >>2884821 pcb is for a low power dc-dc single switch forward converter.

File: Capture.png (275 KB, 1725x1216)

275 KB PNG

>>2884911

>>2884911
>>2884912
No thermal reliefs is a brave choice. What was transformer selection like? Or are you winding your own? I've always been repelled by isolated converters because of that, but if I understand forward converters are easier than flyback converters since you don't need the air-gap in your core.

Also are you sure that optocoupler line should be going into COMP and not FB? The datasheet isn't great, but otherwise it does look like a nice general-purpose current-mode PWM controller, though maybe not quite as versatile as the voltage-mode TL494 with its two error amplifiers.

File: Capture.png (40 KB, 899x609)

40 KB PNG

>>2884925
i designed it with crap i had laying around so it doesn't really have a use case, it's more of just a proof of concept. i believe, however, that it should be good for 48V @ 0.5A output -- meaning that the average current on the caps, didoes, transformer, inductor, and FET are within data sheet parameters.

transformer is a chinky ee25 pc40 ferrite that im going to wind myself. i dont understand magnetics much but i did find a formula that shows 6T on the primary should be enough to avoid saturation. the magnet wire is just stuff that i had on hand, with enough parallel strands to land between 3-5 A/mm^2 current density.

the output inductor was also just some chinky thing i bought off aliexpress years ago for a project i never built. it's rated to """7 amps""" so it should be fine.

i actually wanted to do a flyback but i was too dumb to find/understand the formula for the air gap. single switch forward converter is comparatively simple. i believe my treatment of the COMP and FB pins is appropriate. i also simulated it in ltspice (using an lt1245) and it worked fine.

>>2884926
Just that COMP looks like the low impedance output of an op-amp, I can't see how an opto tied to it could pull it down at all.

File: www.alibaba.com.jpg (2.05 MB, 1115x7530)

2.05 MB JPG

>>2883138
>900M series soldering Iron tips
Recently purchased YIHUA 948-II Soldering Station.

Now I need soldering iron tips..
One of each size of every worthwhile type.
>Pure Copper?
>Gold Plated?
>Lead-free brass material gold plated?
>Lead-free brass material black plated?
!! WTF !!

Recommendations, Please.

>>2884971
You actually want iron plated

File: 1000019008.jpg (456 KB, 1440x1440)

456 KB JPG

Does my buck converter look dead? My soldering iron won't tin properly.

>>2884986

what does your soldering iron tip look like?

File: 1000019009.jpg (294 KB, 1440x1440)

294 KB JPG

>>2884991
Like this. It has a knob and I set it at 350°C. It just turns the solder into balls.

>>2884971
Raw copper ones will probably dissolve pretty fast, you want one with a plating that wets solder decently but doesn't dissolve in common solder alloys. No clue about brass. Lead-free designed tips seem to last longer. Considering the price I'd try one of each. Make sure to get chisel tips though, conicals are ass.
t. T12 chad

>>2884986
Looks fine to me, except for the shitty soldering job.

>>2884992
Is your iron melting solder quickly? If not the temperature adjustment dial may be inaccurate (turn it up), or your thermal conductivity to the tip is bad (consider adding thermal compound inside it). Does your solder have flux in it (i.e. smokes when molten)? If not, get some decent rosin-core solder, leaded if possible. If so, does it wet to the tip (i.e. tinning like you said)? If not, try some tip tinning compound, extra flux, lowering your temperature, or a replacement tip (I recommend a 1.6-2.4mm chisel tip for through-hole work). The tip doesn't look oxidised at all, so I'd sooner guess it to be an issue of flux. If it does wet, it's probably just a matter of technique.

You want to maximise thermal contact with the workpiece, often that means putting your tip on its side if its conical. You should be making the solder joint fast enough that the flux inside the solder doesn't have time to fully burn away, hence leaving you a shiny and well-filleted joint. If this takes too long, you aren't putting enough heat into the joint. This can be improved by tip geometry or angle, by upping the iron's power rating, and in the short-term by using a higher temperature or using a tip with more thermal mass. It may be hampered by a workpiece that itself has a high heat capacity. I recommend having some molten solder stuck to your tip because molten solder makes much better thermal contact compared to a solid tip against a flat circuit board, but you'll still need to add more solder in order to get its fresh flux.

Watch some soldering tutorials if you're unsure.

File: painnnn.png (38 KB, 808x647)

38 KB PNG

>22$ shipping for 47$ worth of parts
It hurts bro.
These SRAM aren’t available locally for me or LCSC so had to order from one of the big boys distributors (Arrow Electronics)

File: 4_c1b98c50-e3b0-42e4-bfbc(...).png (88 KB, 468x375)

88 KB PNG

I wanna splice a computer fan to a USB connector. Do I just solder the +5V and ground lines and it will work if I plug it in.

File: 1732984338374584.png (26 KB, 1014x502)

26 KB PNG

Learning about the dual MOSFET level translator circuit. The circuit seems to work with just the left transistor. Is the second transistor supposed to make the thing more stable? Can I somehow use NPN transistors instead of MOSFETs if I need just unidirectional translation?

>>2885007
>it will work if I plug it in

yes, kinda, maybe
computer fans are almost always 12V
with a 5V supply they'll turn very weakly, if at all
but, yes, 2 wires is all you need

>>2885054
The single FET logic level shifter requires that the higher of the two voltages is on one particular side of the circuit. By having two back-to-back like that, it will operate regardless of if one voltage is higher than another.
NPNs don't work.

File: Untitled-1.png (181 KB, 1595x809)

181 KB PNG

Repairing (again) an karman hardon 370, the two output transistors in pic related had shorted C-E making a short circuit been +VCC(50V) and -VCC(50V). Is there anything else i should check before replacing them and how important are the characteristics of the transistors i replace them with? Can i choose any good ol audio power transistor or does it have to be a 100% match?

>>2885077
You should be able to use any old audio power transistors, if they can handle the voltage and power dissipation, so long as you properly adjust the Vbe multiplier to have the right current through those 0.47R resistors. The +/- 1.2V measurement is likely not something to trust though, what matters is the bias current through the transistors with no signal. You’ll want to tune it down to a low value before putting the new transistors in place.

Figured I'd ask here because /g/ is a bunch of clueless fags:

I recently put some new switches in my Logitech G2023 that started doubleclicking and shit. I replaced the shitty chinese Omron switches with better jap ones.
However, only after I had already put everything back together, I realized that while I was at it I could also have gotten silent switches.
Since I doubt that mouse switch technology will yield new breakthrough technology in the next ten or so years, I would already go looking for good silent switches. Googling is useless, as is googling reddit results, and asking chatgpt got me chinese switches recommended when I was explicitly asking for excluding chinese ones.
Now I am wondering: Are there even non-chinese good silent switches or am I just shit out of luck?

File: boss.png (127 KB, 1042x848)

127 KB PNG

I'm working on a boss amplifier. The dip mounted micro fuse f1 is toast. It oversees the digital supply. D22, D23 test nominal. I've removed C177 and there's only 2 ohm across its unpopulated pads. I'm hoping there's a way to sus this out without getting into the smt part of the board.

>>2885115
No clue about silent switches. Maybe there are hall-effect or optical microswitches in the same form-factor, but I think they'd need another power line. I'd browse Omron's switch catalogue on their website, and download or make a spreadsheet with switches filtered by form-factor, cycle-life, and decibels. Maybe also actuation force.

>>2885117
Yeah 2Ω means there's something shorting somewhere down the line past F1. Short of any smoking guns (or smoking transistors) I'd make a 100mA milliohm meter (literally just an LM317 + a resistor + a 9V battery + a clipping diode) and use it with your multimeter to figure out the source of that 2Ω resistance.

The schematic is kinda confusing with all those "UnPop" labels, is it completely missing that zener darlington regulator circuit?

File: D1G3BqPdA3Q-thumb.jpg (76 KB, 1280x720)

76 KB JPG

I'm looking for something like a magnetic latching solenoid to make a hinge that can switch between swinging freely and locking in place.

>>2885164
From the brief research I did with solenoid actuators, it's cheaper and more flexible to make them yourself. That or you just use a rotary motor and a screw or belt/rack to get linear motion.

>>2885115
there's a lot of autism for mechanical keyboards, i went with gateron zero degree as a decent balance between responsive typing feel and low noise, but i haven't heard of swapping mouse switches being a thing

>>2884765
To remove the shutter knob and slide the casing off, start by examining the knob closely for any visible screws, clips, or fasteners that may be holding it in place. If there are no visible fasteners, it’s possible that the knob is glued on. You can test this by gently wiggling the knob to see if it moves at all, which would indicate it’s likely adhesive. If you suspect glue, carefully apply heat using a hairdryer or heat gun on low for 2-3 minutes to soften the adhesive. Be cautious not to overheat the area, as excessive heat can damage the material. Once the glue has softened, use a plastic pry tool or a flathead screwdriver (with the tip wrapped in tape to avoid scratching) to gently pry the knob off. After the knob is removed, check for any other fasteners or clips that might be securing the casing. If all is clear, the casing should slide off. If the knob is still resistant, check for hidden screws or clips before proceeding further.

>>2885115
> omron switch
I have to replace the switches every year or so in my logitech. I think I’m going to take apart the switch and try and figure out what a actually goes wrong with it.
My guess is that the contacts oxidize as part of their planned obsolescence programme.

>>2883947
usually you have a generic footprint and the symbol assigns the label to pin number

looking to purchase a cheap starter oscilloscope, any recommendations?

>>2885371
What are you going to use it for? What frequency range?

hey

File: 220px-Zebra_connector.jpg (9 KB, 220x193)

9 KB JPG

Has anyone tried replacing Zebra strips (or Elastomeric connectors) in a watch?
I have an old forester "fish in time" from like 2000 and i think the strip has degraded.
How do i make sure i get the correct sizes? (is there a size identifier/number or something?)

File: IMG_2827.jpg (654 KB, 794x847)

654 KB JPG

Manufacturer was very specific about not vaping with this, if I ignore the other suggestions I should be good to charge it with a circuit from a vape tho right. Just gonna be .1/C rate yeah?

>>2885371
dso150 if only for having a convenient handheld scope under $30. Then use your experience to pick a bench scope that will best suit your needs.

>>2885446
>so many "vapers" have killed themselves with this before they died from cancer that the manufacturer saw a need to print this on their products

File: circuitboard.png (75 KB, 821x288)

75 KB PNG

I'm trying to clone a digital out section of a synth expansion board. I'm pretty sure i can source all the parts, but I'm not sure what spec transformer to use for t1. anyone have any ideas?

>>2885572
Not sure, it doesn't say in the datasheet or appnote for my PCM2900s. Look for some S/PDIF appnotes, you should find something.

>>2885446
>Manufacturer was very specific about not vaping with this
It's an unprotected cell.
>>2885572
Pulse transformer
https://www.instructables.com/Improve-Your-Soundcards-SPDIF-Output/

>>2885520
Interesting that they put the onus on battery manufacturers… probably get minor burns on their hands that are covered under healthcare, but the lung cancer will bankrupt you, as is their goal.

> Just gonna be .1/C rate yeah
Trickle charging is generally better for the battery, yes

>>2885572
> transformer?
I’m guessing it’s a pulse isolation/transformer like you see in telephone equipment and ethernet

>>2885572
Any suitably high inductance transformer should work.
You want a coil with significant (i.e. >10kohm) impedence at your audio frequencies of interest, and for audio usually these are sold as "audio transformers" for $1-3 apiece. Ebay has them.
They're the same kind you'd use for a ring mod pedal.

>>2885641
They make ring-modulator pedals? What kind of sound do they produce? Are they used with an LFO for tremolo? Or do they self-mix for harmonic doubling? Seems to me like there’s easier ways of making a VCA, albeit not as linear.

File: 1000019065.jpg (241 KB, 1440x1440)

241 KB JPG

My tips won't get cleaned at all no matter what. This is the 2nd set of tips that won't get tinned. I've never used them, followed tutorials step by step.

>>2885752
>followed tutorials step by step

cant make silk out of sow's ear
those tutorials assume you have a decent tip, like Weller or Hakko
meaning tips manufactured by those companies, not just ''compatible with''

>>2885758
Do hobbyist beginners who watch those videos actually have that kind of equipment laying around

>>2885761
No, your first time hobbyist usually has multiple expensive soldering stations with solid 24k gold 2 oz. tips.

>>2885752
link to where that came from.

File: Capture.png (231 KB, 1018x493)

231 KB PNG

>>2883138
if anyone is looking for some cheap dmm probes i just got these and they're pretty nice. wire is flexible silicone and the tips are sharp and replaceable. definitely better than the crap my uni-t meter came with. they aren't as nice as my probemaster probes but these are like 1/5 the price.

https://www.aliexpress.us/item/2251832765507965.html

>>2885799
Nice replaceable tips, but I miss the days when you could replace the banana plugs and probe bodies with set screws. These look moulded together.
The lead wires are probably PVC which ain’t great, but thanks to the RC crowd, you can get silicone wire that is much better.
Also, I’m definitely going to loose those extra tips, by the time I need them, just like I lost the caps/tip covers.

>>2885844
>but I miss the days when you could replace the banana plugs and probe bodies with set screws.
just buy loose banana plugs and some nice wire

File: banana with set screw.jpg (53 KB, 1024x1024)

53 KB JPG

>>2885844
>banana plugs and probe bodies with set screws

yeah, loved those set screws
so many colleagues electrocuted themselves while unplugging them while touching the screws
it's a lot easier to advance in a company when your competition is dead

>>2885752
Do you tin them when you first use them? Flux them and wrap them with solder and let that burn off. Otherwise they tend to caramelize in short order.

>>2885671
You can add whatever you want lol
The basic effect is just two center tapped audio transformers and 4 diodes. A bit pricier than average, but dead simple.

>>2885752
You're not supposed to clean your tips, you're supposed to leave them covered in fresh tin always

File: a1.jpg (203 KB, 937x1000)

203 KB JPG

What experiment can I do to determine how the drain-source resistance of a JFET varies with the gate voltage? The specific JFET in the voltage controlled filter of this circuit is no longer available as through-hole, so I'm trying to replace it with another one. I need to work out what range of gate voltages will appreciably change its resistance to modulate the filter properties. Would it be as simple as using the experiment circuit that I've drawn? I understand that they are "non ohmic" components and actually voltage-controlled current sources so there's probably more complexity than I appreciate.

>>2885923
Sawtooth at the gate, or just use a potentiometer and draw points.
>using the experiment circuit that I've drawn
Just make a basic bitch common drain amplifier or something

>>2885752
You probably have shitty solder with shitty flux in it, or no flux at all. Does the solder at least melt? See:>>2884995.

>>2885923
> rds
It’s hard to say how sensitive this circuit is to that resistance.
I’d put a transistor socket there and try a few and see what the in-circuit results are like.

> no thru hole.
Solder some wires on to the SOT23.
Get used to it.

File: photo_2025-01-04_18-38-13.jpg (107 KB, 960x1280)

107 KB JPG

https://www.aliexpress.us/item/3256805384870695.html

Hello, I am hoping someone an answer or a possible lead on this. I bought this bldc vacuum motor to be used with a cyclone vacuum for my work bench. I didn't notice initially that you have to cycle to turn on / low speed / high speed / off

pic related is pinout on the other side of the board of where the connector is, you simply short the two wires together to cycle through. I know "U","V","W" is commonly used on pinouts for 3 phase bldc motor, are the ones in pic related have any significance to be able to ID what they are, I mean besides "G".

The wires I short together, one is to ground while the other goes to pin "E"

some other notes that may not matter. Under the board there is 3 blade connectors that I have found to be U,V,W, so if worse comes to worse I will see if I can remove it and use a drone 30A ESC. I mainly wanted to see if I could use a servo tester to adjust its speed like I can with 30A ESC, so I can turn the motor on from the servo tester, adjust speed, and turn it off.

I haven't tried shorting the other pins to ground, and probably won't just so I don't risk destroying the motor.

>>2885752
Use an abrasive like scotch brite to clear off the oxidation, and then immediately turn the iron on, and apply flux and then solder quickly.

>>2885960
No clue sorry anon. If those pads are meant for buttons, then there should be a resistance from them to a positive logic rail, which you might be able to measure when off, or measure the current from them while on. If this is the case, then trying to short different pins to ground (momentarily or constantly) would be a how I'd test it. If not, you could try reverse engineering the PCB by following the traces, with any luck the design is not all in the firmware. Else, yeah you'll be stuck chucking an ESC in there.

>>2885979
Scotch brite is a shitty idea. If there's so little oxidation that a quick wipe with an abrasive will clear it off, then any half-active flux will do that with less damage to the plating. Plus some rubbing against something soft while it's hot, like a wad of paper towel. If not, then you're even more likely to ruin the tip with the abrasive, and should probably use a dedicated tip tinning compound instead. Maybe zinc chloride or molten citric acid would work too, idk.

File: 1509040234090.png (176 KB, 504x337)

176 KB PNG

>>2885923 (me)
In the end I managed to find new values for R43, 44 and 45 by replacing them with variable resistors and tweaking them until they worked. I wish I could be more sophisticated, but I suppose an experimental approach is OK if it gets you results.
I'm kind of concerned about this JFET going out of production though, are all through-hole components eventually going to be replaced with surface mount? How will hobby electronics survive?

File: 4C251693-C4E8-43FE-BC65-9(...).jpg (302 KB, 1389x533)

302 KB JPG

>>2886222
Heh heh, I heard the same arguments about being impossible to repair and whatnot when p2p was being phased out in favour of PCBs: “cheap crap so they can fire all the assembly line shifts”

Turns out it ushered in what most people think of as the golden age of hobbiest electronics.

>>2886222
JFETs specifically are pretty bad for going out of production. Especially since their circuits are even more dependant on their turn-on voltage than a BJT circuit is on hFE. It’s usually pretty easy to just make or buy an SMD breakout board though.

I’m an op-amp guy, partially for this reason. If that op-amp is hard to get, just swap another in with equal or better specs where it matters. It gets more difficult and more expensive when it comes to RF circuits, but if you’re sticking to audio stuff, you can’t go wrong with op-amps.

File: 456966883_101716665708130(...).jpg (43 KB, 526x592)

43 KB JPG

>>2883138
Once upon a time there was a list of 100 /ohm/ projects to roll floating around. Does any of you have it?

Also, any projects for a complete beginner? I don't want to play around with arduino and build robots, I want to get a good grip on building functional circuits with basic components (resistors, capacitors, etc)

File: 1600.jpg (85 KB, 1600x1600)

85 KB JPG

>>2883138
Yo there! Good year!
Sorry to bother again, I just installed a new led light like pic rel, with the same kind of driver, I make the couples of the wires and put in the socked, but it started to flicker, in the same way when the caps are damaged, changed the driver and still do it, changed the led light and still does it, cut the round connector and tie the wires, but still flicker.
I changed another new one with their own driver and still flickers.
The first light was used for a couple of days with no problem at all, in another circuit, but still in the same house; could it be something about the electrical installation?
When starting to make the fault, it stopped when rotating the plug, so I was guessing it
was a false contact, since if the driver went fuse, it would keep that state not going back and forth, and to fail all the drivers/lamps, well, while I think its possible I don't think it is the case.
What's wrong?What can be done to correct it?No other bulb,lights,devices shows fails in the same circuit.
Thanks.

>>2886323
>Does any of you have it?

go to link in first post: https://github.com/74HC14/ohmOP
search ''dont ask''

File: 1573260370298.jpg (48 KB, 680x453)

48 KB JPG

>>2886282
That's a pain that JFETs are falling out of favour, a lot of analogue synthesizer circuits seem to use them, and those are what I'm interested in building.
I'm experiencing this dependency on turn-on voltage firsthand because the circuit I'm making turns out to be extremely sensitive to the gate-source voltage on Q8, which is itself dependent on the output of another JFET.....

>>2886355
As was said, you can re-bias the gate with a voltage divider, using trimpots to adjust it. Then, take out those trimpots and measure them, replace with resistors.

I have no idea what the open loop gain of an actual op-amp is, it probably varies with the hfe of several transistors multiplied together…. It’s gonna be a lot.
That’s why production circuits using BJTs or op-amps use techniques (such as negative feedback) to eliminate a lot of the individual device variances.
It’s expensive to have some guy tweak a potentiometer in a test jig on an assembly line.

For personal, one-off circuits, by all means, put in those trimpots and set them depending on the specific hfe or gain of the device. You may have to do it again if you have to replace the transistor—even with the same part number—due to individual device characteristics if the circuit is sensitive to the exact parameters.
Things like the 2N3055 went through several entire semiconductor process changes (and manufacturers) which changed things significantly back in the day.
Nowadays, you probably have to deal with chinese buying bad batches and re-selling them.

>>2886355
>JFETs are falling out of favour
It's more that THT JFETs are falling out of favour. There may be more JFETs now than 30 years ago, at least if you include chinky chips you can buy on LCSC that's probably true, but they're mostly SMTs. There's definitely less part turnover with JFETs compared to power MOSFETs, which are constantly improved upon with only the old jellybeans remaining, but it's probably more in flux compared to BJTs. If you're designing a JFET circuit, use jellybeans since they'll probably be around for a while. But if you've got a JFET circuit where the engineer chose a JFET that was cutting-edge for its day, you'll probably be unable to find an exact replacement, at least not without sketchy ebay listings.

>>2886368
>to eliminate a lot of the individual device variances
Not just batch variations, but changing gain as a function of temperature or device wear also gets minimised thanks to negative feedback. If you use topologies like chopper-amplifiers, you can negate the input voltage offset, regardless of how well matched the input stage is.

Also JFET-input op-amps are probably cheaper than an equivalent discrete JFET amplifier circuit. Not that this applies if you're using JFETs as voltage-controlled resistors, like in a VCA or mixer. Operational transconductance amplifiers do exist, but IIRC they're not ideal and don't have the feedback benefits that conventional op-amps do. Dedicated mixer ICs are bad for general-purpose circuitry.

File: 54hjww.jpg (57 KB, 607x794)

57 KB JPG

>>2883138
Working on a home project, tiny wee 3-axes home cnc for hobbyist jobs, Aircraft frames and the like - from low density woods, foams and ideally acrylic, Having trouble with the spindle motor and power supply. Ideally this would be as dirt cheap as possible. looking for options as to what I can do. Do I make the power supply? would it be worth trying to make my own motor?

Less than half a meter on all sides. Aluminium extrusion, lead screws, nema17 motors, arduino uno 3 w/t drivers. i am a tight fisted bastard so act accordingly

File: a4ebff4f-1675-4c70-a3b0-f(...).jpg (49 KB, 466x700)

49 KB JPG

so i bought this shitter for very cheap, used, i like its big display and that it reads caps, it does everything except reading voltages of any kind. already changed fuse (which was cut) to no avail. any ideas what could it be? nothing fried under pcb under simple inspection.

>>2885923
that fet could be literally exchanged for a LED, or a bjt, its very uncritical for what its doing there. id suggest maybe do a diff project? do the oberheim SEM (but put in 3340 vcos)

>>2886436
For the spindle motor, it depends on if the motor is static or if the workpiece is static. A static motor could be any beefy mains motor or a treadmill motor or whatever, weight being a bonus. A static workpiece means the motor must be lightweight enough to move about, so you’re probably looking at some sorta PMSM or BLDC.

For the power supply, just chain a couple 2nd hand ATX supplies together to get 24V. More if you want a higher voltage. Older supplies do more power out their 12V rails, newer supplies focus on their 5V and 3.3V rails. You will need to remove the ground reference on all but one of them, I did this via a plastic washer between the PCB and chassis. The more that runs on mains voltage, the less of a power supply you need.

>>2886571
>BJT
No it couldn’t be replaced with a BJT. The signal it’s pulling down is centred about ground, a BJT only allows current in one direction. That JFET is being used as a voltage controlled resistance.
>LED
what the fuck are you on about?

>>2886436
>Do I make the power supply? would it be worth trying to make my own motor?

Fuck no. Ready-to-go spindle motors can be had dirt cheap from everyone favorite communist utopia.

https://www.ebay.com/itm/375627355237

I guarantee the runout and vibration characteristics are trash, but not to the point that it should cause problems with wood, plastic, and foam. Find or hack together a suitable power supply and that's probably about the best you're going to get on the cheap.

File: Busted.png (268 KB, 1513x753)

268 KB PNG

Was buying picrel a mistake? I'm building a steam distiller and need to record pressure. I'm trying to do so with an Arduino but this thing doesn't seem to output any consistent voltages/currents at all.
I've got it hooked up to a 1.0 A / 12V power supply, I'm running the signal wire through a 250 Ohm resistor (to achieve a max of 5.0 V), and feeding that to the A0 pin. It will sit at either 0.0V or 5.0V for a few seconds before quickly switching to the other extreme.
My only real requirements are that (1) it measures pressure, (2) it can report to an Arduino, and (3) has a 1/4" NPT adapter

>>2886590

Is it _actually_ swinging around in voltage, as measured by a multimeter/scope, or is that just what you're seeing in the serial output? If the former is happening, and you're sure the power supply is good, the sensor is likely defective. If the latter, it's a code issue.

File: 1000019100.jpg (416 KB, 1440x1440)

416 KB JPG

How can I find a compatible LCD for this controller? I can't find the exact LCD from the one that broke but I'm wondering if I could use a different perhaps smaller screen instead. On google, I found similar to the one I have. https://m.globalsources.com/LCD-controller/VGA-HDMI-of-LCD-Control-Board-technic-support-1200694756p.htm

>>2886594
I don't have a multimeter and am going only off of serial output. I thought taking the value from A0 would be enough, which itself is fluctuating, but that may not be the case?

>>2886697
>I don't have a multimeter

you can use your own ears, or those of an intimate friend, as a signal detector up to, maybe, 15Khz
send data pin to your stereo or computer speaker by using a series 10K resistor and a small cap
if it's chirpy-chirp cherping, it's cereal data

>>2886590
Is this thing even supposed to be compatible with arduino? You might find that it was designed to work with some kind of dedicated meter

>>2886737
>some kind of dedicated meter

title says it's a common 4-20mA analog loop
https://www.predig.com/indicatorpage/back-basics-fundamentals-4-20-ma-current-loops

>>2886590
Is the sensor's output voltage listed? Do you need a pullup or pulldown resistor on its output? Maybe a 4.7V zener + resistor.

File: 4-20mA_current_loop_components.jpg (104 KB, 1024x512)

104 KB JPG

>>2886740
This page says the signal is encoded in current rather than voltage.
> It will sit at either 0.0V or 5.0V for a few seconds before quickly switching to the other extreme.
You should probably focus on measuring and interpreting the current.

>>2886742
No, I had to infer it from the output amps. Would running a 250 ohm resistor on the 20mA max not yield the expected voltage of 5V for the Arduino?

>>2886744
Very interesting. So >>2886745 is false. Do I need dedicated hardware to interpret current? I don't see a path for doing this with my current setup. Thanks for the info.

>>2886747
If you're really bent on using arduino, you could probably find some kind of chip that will interpret it and send it to arduino via SPI/I2C, or maybe find a hack to do it with arduino + couple of passive components/arduino itself.

>Do I need dedicated hardware to interpret current?
Just to be clear, by dedicated hardware I mean a general type of pressure meter or a display that adheres to 4-20 loop standard. They don't have to come from the same brand, since the standard kind of implies they're supposed to be interchangable - that's one of the reasons why the sensor is sold without the meter.

>>2886757
Would the ACS712 Current Sensor achieve the function you're describing?

File: Transimpedance-Amplifier-(...).png (29 KB, 750x474)

29 KB PNG

What's the point of LCSC when the shipping is so expensive?

>>2886747
I found this. No idea if it's any more helpful than the other link.
https://hackaday.com/2017/07/19/the-4-20-ma-current-loop/

You could try building a TIA. Maybe your sensor doesn't like the shunt resistance you're using or something.

>>2886763
It would measure the current up to 0.5A and send it to arduino, but even if you get the current readings, your arduino still has to interpret the signal you're getting, so unless you find a library/script that's designed to work with pressure meters (suprisingly, there might be one out there), you might have to run a couple of tests and write the code yourself.

espressive states the esp32 analog input can measure between 150 mV ~ 2450 mV in max range. What would happen if i overshoot and put lets say 2.6 volt on the pin? fry it or simply output max value?
cause i have a 0-5v sensor and im thinking the best way of hooking that up is a simple voltage divider

>>2886747
>Do I need dedicated hardware to interpret current?

a resistor does a good job of converting current to voltage
but i'm guessing you wired it in series, which is wrong
you need to sample the voltage *across* the resistor
if the arduino sucks in too much current, that'll affect accuracy
in such a case, you'd need an op-amp buffer, capable of rail-to-rail outputs, or powered from, say +-10V

File: cringe stniffing dog doge.jpg (48 KB, 480x597)

48 KB JPG

>>2886697
>I don't have a multimeter
>pic related

Stop spending money on food and water until you get one, this is not an optional thing for electronics. It doesn't have to be good, it just has to be vaguely accurate.

Ignore the others giving you weird advise, I'm not sure they know much more than you do. Just to clarify, what you bought is a sensor module with a standard 4-20mA output signal, 4mA being the signal it puts out at the bottom if its measuring range, and 20mA at the top. Anything below 4mA should be interpreted as an error signal, the most obvious problem being "it's not connected". It is not "compatible with Arduino" specifically, in the sense that it it doesn't come with a USB stick containing an entire configuration library and plug-and-play shield. But it's trivially easy (or should be) to use, regardless.

>>2886742
>Is the sensor's output voltage listed?

It outputs *current* as a signal, not voltage. It will put out whatever voltage necessary to produce the current it wants to, up to whatever maximum voltage it can (likely Vcc). The easiest way to measure this signal is to pass it to ground through a resistor, and measure the voltage drop across that resistor. I assume >>2886590 has figure that out, given that 20mA through a 250Ω resistor produces a 5V drop. That signal can then be sampled by the Tarduino's ADC, and the value received can be interpreted as you wish.

Without measuring the actual signal, it's hard to say what's going on. Could easily be the ADC is just being misconfigured. Depending on how this was wired up and what the sensor can put out in terms of voltage, part of the microcontroller on the Arduino may have also gotten cooked.

>>2886795
You want to find the absolute maximum ratings table. If they don’t have one, they’re not really legitimate.

If you keep the current under, say, 5mA, then 3 1N4148 signal diodes in series placed across the input will limit it to 2.1 V. Alternately, you could use something like an InGaAlP yellow LED with a Vf of 2.3 V or some series combination of various crap from your parts bin. Don’t forget to test the actual total Vf of whatever you come up with.

>>2886582
>what the fuck are you on about?
things you dont know, tard. both LED and bjt would work well, but thats such shitty ass filter that nobody should bother with it.

File: ba.jpg (349 KB, 1323x849)

349 KB JPG

What's the easiest way to interface potentiometers with a breadboard? Currently I solder single core wire onto them and have them sticking up in the air like this, but it gets cumbersome to use. I know that there are "trimmer pots" that could be inserted like through-hole components, but I also understand that they generally aren't rated for many turns and would wear out fast.

>>2886814
Do you perhaps mean a vactrol? Vactrols are kinda slow, but probably fine if your LFO is slow enough. They aren't that linear, but you can probably fix that with some feedback and a second vactrol.

A BJT definitely wouldn't work, it would clip half the waveform. Provide a schematic or counter argument instead of "no u".

>>2886907
Consider 9mm pots, they fit decently into breadboard rows.

>>2886772
>What's the point of LCSC when the shipping is so expensive?
You have to buy at least $50 worth of shit to make it worth it. Probably makes more sense to people living closer to China.

>>2886795
I might be wrong, but generally ADC inputs can go as high as the analog VCC rail. The ESP32's voltage rails appear to all be at 3.3V. The 2.45V measurement limit will just be because that's as high as the ADC reference voltage goes. So you've got a good 0.8V of headroom to fit in a 3V zener or something similar.

>>2886582
>>2886588

Thank you, gentlemen.

You are both truly wonderful and I am forever in your debt.

Goodbye forever,
Grael

>>2886802
Nice facebook meme.

File: CD77ACE9-5C94-4D61-A5B2-6(...).jpg (1.5 MB, 1000x1617)

1.5 MB JPG

>>2886907
Change your breadboard construction technique, make it easier on yourself instead of building a bird’s nest.

>>2886907
Mount the pots to a piece of perfboard wired to a pin header and connect them to the breadboard either directly by pushing the pins into the breadboard or with a ribbon cable/dupont jumpers.

File: pots on your breadboard.jpg (164 KB, 585x800)

164 KB JPG

>>2886907
>interface potentiometers with a breadboard?

- build an outboard jig with many pots, switches, LEDs 'n shit
- build lil PCB jigs for each pot
- simplest one: cut a piece of 1mm-thick plastic sheet, add cutouts that fit into nipples on side of breadboard
. drill holes in said sheet for pots, switches, LEDs 'n shit
. attach plastic sheet to side of breadboard by jamming said cutout into said nipples

>>2887142
>1mm-thick plastic sheet

forgot the best part
- make 3D printed ones and sell on etsy for 1 million pesos' worth

File: 2046.jpg (328 KB, 1000x804)

328 KB JPG

Sup, I've got a Yamaha VSS100, sampling keyboard. Every time i do a sampling, if i keep the key pressed, once it stops sample playback it immediately starts a loud squarewave that keeps going for a total of 8 seconds (sample+squarewave time = 8sec). 8 secs is maximum sampling time as per manual. Could this be indicator of RAM chip gone awry? Its a MB81256-15. It has shown also memory perseverance, like it recalls part of a past recorded sample, even after being turned off and on, it really shouldnt.

File: out.webm (1.41 MB, 270x480)

1.41 MB WEBM

>>2886814
>such shitty ass filter that nobody should bother with it.
I don't know what it's doing to my square wave but it's sure as heck not filtering it. I probably built it wrong somehow. Sweeping R28 through its full range gives me a square wave with those resonance peaks at either extreme (so some filtering is happening) but anywhere in the middle, my square wave is for some reason being converted to a sine wave. The guide for this circuit calls it a "T-type filter", but searching that online just gives me water filters - does anyone know something relevant I can read to understand how it's supposed to work?

>>2887084
I don't have time to be so neat I am a busy man

>>2887124
>>2887142
Thanks for the suggestions, that's a fair bit of work, but I suppose it would be useful to make up some "generic" control panel that can be plugged in to arbitrary breadboards...

>>2887174
T and pi filters are common topologies for filters. For a low-pass T filter, you'd usually have two series resistors, with a capacitor between them to ground. A lossless filter would use inductors instead of resistors, and an active filter may use virtual inductors instead of resistors. That said, usually you'd only use a 3-element pi or T filter instead of a conventional two-element filter if you're feeding the signal into a non-trivial impedance.
A standard butterworth low-pass filter would just have a flat pass-band that smoothly transitions into an n* -20db/dec rolloff, where n is the number of poles of the filter. But a filter with inductive and capacitive components will often have a resonant peak between those two sections, the amplitude of which is determined by the Q factor of the filter. In your case it definitely looks like there's such an element as you adjust the cutoff frequency of the filter, but it's hard to say if the cutoff frequency is all that's varying. It looks more like a wah-like filter, where you're moving a resonant band-pass peak back and forth among a flat finite gain background, otherwise it doesn't really make sense for there to be a square wave on both ends of the filter's voltage range.

I'd simulate it.

File: Zrzut ekranu 2025-01-09 055315.png (35 KB, 430x410)

35 KB PNG

>>2883138
What is the purpose of a resistor in RC high pass filter?

>>2887185
>What

Fc = 1 / 2 pi R C
without the R, the formula goes to zero

>>2887185
To set the cutoff frequency, 2*pi*f = 1/(R*C). Basically that's the point where the resistance is equal to the magnitude of the capacitive impedance.
In an ideal case, with an infinite load impedance after the filter, the full signal will get through that capacitor, even at low frequencies. In a real case, you may be feeding an op-amp or transistor circuit, where the load impedance of this circuit may vary as a function of voltage or other circuit parameter, meaning the cutoff frequency will be shifting constantly.

Furthermore, the resistor provides a DC reference. If you had a high frequency sine wave on the input of the filter that's referenced to ground, its DC level after the filter could drift around in the absence of load current, or be preferentially pulled high or low by asymmetric load currents. Conventionally you'd use a resistor to a ground rail, but if you do that to a single-rail system than half of your signal is going to be outside of your supply rails. A guitar pedal will often use a single supply voltage, where the input signal is referenced to ground, so often they use two resistors on their high-pass filter, one to + and one to GND, to pull the signal between the rails. On the other hand, a charge pump circuit deliberately leverages an AC wave going out of the rails to increase voltage or invert rails, like you see in a high-side gate drive IC or MAX232.

File: white-leviton-outlets-rec(...).jpg (49 KB, 1000x1000)

49 KB JPG

if i stick a normal plug to the 20A receptacle, is it still capable of drawing ~20A? or is there some kind of internal mechanism that limits output if the T-side plug isn't horizontal?

>>2887207
It could. It's no different than sticking any other device into a 15A plug. It's up to the device to ensure proper fusing/overcurrent protection. There's no way for a mains plug to limit current. It's just a simple set of contacts. Even things like GFCI outlets can only shut off power.

>>2887207
The assumption is always that the device being plugged in is safe. If you have an appliance with a 10A plug that draws 20A, that's your problem. The wall outlets are not designed to fail safe in that event. That said, the spring contacts inside the live and neutral pin sockets may well be stronger for 20A sockets, in order to carry current with less contact loss.

>>2887210
Well 10A plugs aren't fused at 10A in the first place, you have multiple plugs going back to one breaker. In my house they're on 20A breakers. So there's nothing stopping you from drawing 20A from a 10A socket, say if you put a 10A plug on a welder or whatever.

Though considering $5 powerstrips have 10A breakers inside them, I'm surprised we don't just have a 10A breaker on every wall outlet. It would solve the splitter plug problem. Put that bimetallic strip near the contacts too while you're at it, that way if the contact gets shitty enough to produce too much heat it will give you overtemperature protection too.

>>2887207
No limiting in the outlet, not that it matters. A compliant 15A device will never draw more than that.
If it does draw more, something has gone very wrong and all bets are off. Most likely either a fuse in the device will blow, or it'll exceed 20A anyway and blow the breaker.

>>2887207
>stick a normal plug to the 20A receptacle, is it still capable of drawing ~20A?
It will work but, be cautious...
If the plug & lead isn't designed to handle 20A, then it will eventually melt and catch fire..

>same problem as sticking 20A plug into 15A socket
20A plugs will not fit into a 15A wall socket..
If you modify one of them so the 20A plug will fit into the 15A socket, then it will happily still work until the bits of socket plastic and wiring insulation melts and catches fire.

>>2887239
> melts and catches fire
I doubt this would happen. The horizontal prong is on the neutral side… I imagine the construction is otherwise identical for standard 15 A sockets.

I know lots of people who have 20 A outlets in their kitchen, yet I have never possessed, seen for sale, or know anyone that has a device with this kind of plug over the last 20 years or so when I first saw those start to appear.
I suspect it’s more of an indicator that it’s wired with an unbranched 12 AWG circuit with a 20 A breaker, and you, the homeowner, should prefer using those sockets for things like 1700 W hotplates as less energy will be lost inside the walls.

>>2887185
I suspect, intuitively, without the R, one might see that smaller values of C pass higher frequencies and that larger values of C would pass lower frequencies.

File: sw.png (41 KB, 1019x630)

41 KB PNG

How to implement the switch in this sketch with transistor(s)? The circuit I'm working on needs to separate the op-amp output (circuit A) from another circuit B. There's a diode which blocks the flow of current from B to A in any case.

>>2887270
15A power sockets and wiring insulation melts and catches fire if you run a 20A tool thru them.
It may take a few hours or so, but it happens.

Seen it happen & done it myself.. Be told.

>>2887328
You can't buy a NEMA 5-15R that isn't rated for 20 A pass-through.

>>2887329
melting and fire at the socket starts at the connection hotspot melting the house wiring insulation.

>>2887188
>>2887190
This does make sense, but can you also explain why the resistor is not in series with the capacitor? If the output branches out and doesn't go through the resistor, why is the resistor included in the equation for the output voltage? Or maybe my undestanding on what the output is is wrong

>>2887349
>why the resistor is not in series with the capacitor?

the resistor and capacitor make a voltage divider
the capacitor's resistance (or ''impedance'' as the nerds call it) varies with frequency according to Xc = 1 / 2 pi f C
so as frequency goes up, the impedance of the top element of the voltage divider drops, so more signal goes through
making this arrangement a high-pass filter
obviously, if you reverse the components you get a low-pass (aka, "high-cut" or "high-reject" which is what my mom called me)

>>2887318
Ok first of all, having that diode blocking reverse current being outside of the op-amp's feedback loop means you'll have a ~0.6V drop, so you wouldn't actually get 8.375V on the output at all.

For the switch, I'd probably use a P-channel MOSFET.

>>2887371
Great explanation!
Seems to me the filter will burn a lot of the total incoming waveform off as heat in that resistor though, so no very efficient imho

>>2887428
RC filters are generally not used for power waveforms (e.g. a radio or audio signal after the amplifier but before the antenna/speaker), here you'd typically use an LC filter instead. RC filters are generally used for small-signal processing, like for anti-aliasing before an ADC, or to isolate different frequency bands of an audio signal before the amplifier in a graphic equaliser.

>>2887442
>or to isolate different frequency bands of an audio signal before the amplifier in a graphic equaliser.
As part of an Active-Filter / Gyrator?
An RC circuit wouldn't have the dB per octave selectivity for an audio qraphic equaliser.

>>2887428
You should buffer the signal after the filter, especially if it has multiple stages

>>2886907
cut each of those wires to ~1cm, so the chassis of the pot hangs off the breadboard and the three short wires plug into 3 lanes

>>2886795
Less than a diode drop generally isn't an issue, but of course do everything at your own risk.

File: 1733821977094910.jpg (50 KB, 669x516)

50 KB JPG

>>2887428 >>2887449
>buffer the signal
what Bode Plot Margin are your cascaded amplifier/buffer stages allowing?
How much distortion do you all aim for?

>>2887444
I was just using it as a noob-friendly example for why one might want a filter. I've never made a graphic eq before, and I might have seen only one or two schematics before. I assume you could get half-decent performance from 20db/dec rolloffs if you only had 5 or 6 bands you wanted to control, not like you'd want a stair-step frequency response. Otherwise yeah you'd want something steeper, probably an active filter for each band unless you can get away some funky passive twin-t bullshit.

File: 3 band EQ.jpg (186 KB, 793x769)

186 KB JPG

>>2887490

It would be nice to have a general-purpose CC/CV charge control chip, with end-charge hysteresis. One that could do various chemistries with numerous series cell counts, even high voltage ones if you gave it a divider. One that could output PWM to a switching regulator's FET, or just linearly regulate via a BJT. Just configure it with like 3 resistors. Such a chip probably exists somewhere in the sea of sino-silicon within LCSC.

>>2887442
>>2887449
>>2887481
Seems to me, you might want to amplify the signal both before and after the RC filter.
This, of course, is going to start introducing a lot of noise from all the amplification.
I’m guessing there is some kind of “conservation of filtering ability” theorem where you can’t make a filter that is good in all dimensions.

I wanna specifically start a robotics general guys. What should I put in it?

File: 1724709128661346.jpg (63 KB, 740x960)

63 KB JPG

>>2883138
bros I just want to make electronic components in my garage and sell them, but I feel like I couldn't compete with already established companies in anything. What could I possibly do at home except for bundles of stripped wires?

File: 1734095417419836.jpg (169 KB, 1080x1439)

169 KB JPG

>>2887497
>Line level Push-Pull bands
CLASS-AB pre-amp!!

>>2887549
>Find the component-level faults in common household electronics and sell repair kits or aftermarket/equivalent parts that are out of production.
>Restore old shit and jack the price up for nostalgiafags on the ebays.
>Print arcade cabinet/control panel/marquee artwork and decals, or back glass art on pinball machines.
>Wind custom transformers and coils.
>Take empty shell casings for different calibers of weapons and put laser diodes in them with coin cell batteries and a switch for sight alignment.

>>2887549
Restore power tools and spot weld battery packs.

>>2887549
Custom-wound indctors and transformers, especially toroids if you make a machine for that

File: variable-capacitor-7e9670(...).jpg (105 KB, 750x591)

105 KB JPG

>>2887549
sell +3500ppm/C platinum PTC resistors for synth repair
or diy variable capacitors, some of those big ganged ones can go for quite a bit

File: IMG_20250110_212831339.jpg (844 KB, 4160x3120)

844 KB JPG

newfag here, not sure if this is there is a dedicated lighting thread so im putting this here, i just ripped an old fluorecent tube fixture out of the wall and was wondering how to wire it up
do i just connect a old discarded plug to the cut ends or do i need o do something more?

>>2887720
>>2887727
That's the ballast but it's missing the starter that goes in the socket by the round hole. You can wire it with a regular appliance plug. Hot to black and neutral to white (the short wire on the left bulb socket). Plug it into a power strip so you can switch it on and off to test it.

>>2887728
thank you dude, i took out the starter bc when i was removing it i wanted to make sure the power was broken to it since my house is 120 years old so shutting off the power to snip the wires was out of the question due to the fusebox's fragility
i also ripped out a ceiling fixture and was wondering the same thing, its different bc i don't see any starter, will include pic next post

File: humidifier pcb.jpg (3.62 MB, 2997x2000)

3.62 MB JPG

Hello i'm a n00b with electronics
I want to hook up a humidifier to a relay to turn it on and off when humidity reaches a certain range

This is the Humidifier PCB in picrelated

I have the relay, a XY-TR01, working properly but the issue I'm having is that these humidifiers have a switch to activate the ultra sonic discs and make them actually spray humidity into the air on their PCBs.

The middle of the top three RC connectors was a battery that I removed so it can be purely USB powered. The one off the the side is for the humidifiers LED lights. The other two go to the discs.

The humidifier has multiple modes so you press it once to turn it on, then you have to press it again to activate the discs. Holding down the switch and turning the humidifier on won't activate the discs so I can't just tape it down.

Is there a way I can bridge any points to bypass the switch? I heard that they make switches now that activate on a pulse instead of just always being on when pressed so I don't know if it's possible with this one.

Also is there anything stopping me from just soldering the positive and negative wires of the discs directly onto the positive and negative pads where the USB plugs into the PCB?

Sorry I am retarded and I don't want to blow myself up

File: IMG_20250110_230822925.jpg (780 KB, 3120x4160)

780 KB JPG

i have wired it up using a plug from my uncles white noise machine with the right wire (goes to wide end) to white and black ballast wire to narrow end wire (left) and installed the starter back in its socket
how do i arm this thing without potentially blowing my ass up?

>>2887743
>how

stop being a chickenshit
that thing's worked for 2 centuries without blowing up, it shouldn't blow now
besides there's nothing there than can blow if it wanted to
put in the tube, hide behind the sofa, and plug it in

BTW, if the other lamp has a big-ass ballast about 8 inches long, those dont need no gay starters

File: IMG_20250110_233140439.jpg (523 KB, 4160x3120)

523 KB JPG

>>2887747
hey man, i might be electronically retarded but i am aware enough to be wary when playing with mains electricity
plus i only have 2 bulbs, since i ripped this fixture out of a bathroom mirror so don't want to have to go all the way to home depot to get more if they do blow so its more that i want to be extra careful so i can be lazy
i have another fixture that has a plug without any type of wide ends (tube) which doesn't seem to care about orientation so does it even matter?

>>2887749
>so does it even matter?

no, regular edison bulbs need to be polarized for safety, but fluorescents dont care
coz they're equally unsafe at both ends of the tube

File: IMG_20250110_234314699.jpg (650 KB, 3120x4160)

650 KB JPG

>>2887743
pic of the other lamp's ballast, its a ceiling circline fixture (t9)
this one might not need a starter bc the diagram on the ballast but i could be wrong

>>2887742
>stopping me from just soldering the positive and negative wires of the discs directly

they need high a frequency signal to get them to vibrate
that's what the black circles on PCB are generating
DC wont work

>>2887754
>might not need a starter

and yet it says "trigger-start ballast" right on the can

>>2887755
i see thank you so is there any way to bridge it passed the switch or should i just look for another board

>>2887760
>is there any way to bridge it

of course there is, but it requires some expertise (and electronics instruments) to find and fix it
after the second keypress, that lil black control chip sends out a signal to start the action
you'd have to find that signal and add a fake ''go'' signal
which is beyond your ken

>>2887765
yeah that does sound way beyond me
could i use a multimeter to find where the signal is sending and solder a wire from there to the usb power? or is it something i cant even conceive of doing?

>>2887767

you might get lucky with a voltmeter, or you could destroy the whole thing
there are 2 chips, U2 and U3 which might be voltage regulators
meaning the circuit might actually work at 3V (or whatever) not 5V
so connecting signals to 5V might be fatal to the chip

>>2887775
>so connecting signals to 5V might be fatal to the chip
i think that would be the case
i had one before this that i tried to wire up in every possible way but eventually it just stopped turning on the discs even after desoldering everything
ill just get a new chip i think at this point
thank you again for your help

>>2887779
seeing what i think are a bunch of resistors on there between the chip and whatever else too seems to suggest that

File: IMG_20250111_001508170.jpg (717 KB, 4160x3120)

717 KB JPG

>>2887758
well im saying that because the diagram on the ballast doesn't seem to show it in the circuit like >>2887727 as well as other videos about circline ceiling fixtures having a built in starter and my memories of turning the light on and it not seeming to preheat like the others i have

File: IMG_20250111_005041225.jpg (806 KB, 4160x3120)

806 KB JPG

SUCCESS!

>>2887700
>+3500ppm/C platinum PTC resistors
I've been reading a lot about synthesizer electronics recently and it seems crazy that so many VCOs rely on this one specific kind of thermistor that nobody makes any more. How can there be no other applications for such a component? Did everyone switch to different methods of temperature compensation? Other thermistors on suppliers' websites never even seem to be described in terms of ppm per degree.

>>2887858
Because the kids these days simulate analog synths with VST plugins on a digital workstation.

File: 1586520783669.png (1.22 MB, 400x800)

1.22 MB PNG

>>2887798
>warm white
>piss white
At least it glows. You should use a 3-prong power cable with the ground wire connected to the metal housing of the ballast if you plan on using them. 2-prong is okay for testing so long as the hot lead isn't exposed and pressed against the housing. lol

>>2887858
Surely there’s an affordable IC from the 90s that has a temperature compensated lin/log converter and VCO ready to go, right? A proper analog one, not some bullshit ADC+DDS chip.

>>2888020
Light bulb

File: jjjj.jpg (233 KB, 784x867)

233 KB JPG

>>2887084
I cleaned up my act and used wire jumpers instead of long flying leads and wow, it really does make a difference.

>>2888020
There are chips like the AS3340 VCO but I think such specialist stuff can be difficult to source

>>2887858
>How can there be no other applications for such a component?
There are tons of applications, +3500ppm/C is meant to offset the -3500ppm/C tempco of a silicon junction

File: 2025-01-12-16-22-27-657.jpg (2.38 MB, 3000x4000)

2.38 MB JPG

Maybe I should clean

File: Screenshot_20250112_17485(...).jpg (468 KB, 1080x1057)

468 KB JPG

>>2888211
>>2888211
I am completely new to electronics in the sense that I've only ever made a pixie morse code radio thing, a little 5 dollar soldering introduction at the amateur radio club.

What i want to be able to do is household appliance electronic repairs.

I have a ham radio license, which I did a course for and it covered basic of electronic theory.

Should I just buy a soldering kit and make projects off Amazon? Or should I legit get Make Electronics book and do all the experiments in it?

If you were starting from complete scratch but wanted to work on this as a hobby after the kids are in bed, how would you start?

>>2888270
Whatever you do, if you want to have the knowledge-base to understand circuits for appliances that don’t have an annotated service manual, you’ll need to have a lot of iterative learning. Expose yourself to a lot of different circuits in an environment where you have a pathway to understanding them. When you see something you don’t understand, find an article or video on the topic. Big Clive is actually pretty good for exposing you to simple Chinese appliance circuits and explaining how they work after reverse engineering their schematics. For building circuits, following a book will give you some foundational understanding, but you can get that kind of understanding just by reading. Less so for the intuition you get from building g the circuits in such a book. Actually designing circuits and having to troubleshoot your own designs is the fast track to learning how to make circuits, so long as you have the required foundational knowledge. Learning how to design novel circuits helps you put yourself in the shoes of the designer when it comes to repair, and also allows you the freedom to substitute or modify part of an existing circuit. Like spoofing a button press, or adding a soft-start feature.

>>2888288
Thanks anon, I'll reread this post many times.
I'm reading the pdf version of make electronics right now. I think I gotta get a decent multimeter and soldering station and a set of components next.

If you wanted to measure the distance between a piece of equipment and a wall or floor that a component on that equipment was pointing at, how would you do it?
What I need here is essentially one pixel of lidar.

File: optotriac board.jpg (2.33 MB, 3622x2789)

2.33 MB JPG

Time to get around to making this hot-plate. I think I've got a thermocouple reading module lying about somewhere.

File: ultrasonic range finders.jpg (145 KB, 1375x614)

145 KB JPG

>>2888346
>measure the distance

ultrasonic range finders
sends a ping, waits for an echo, calculates distance based on speed of sound
which is 55 MPH at 25-deg C, at sea level

>>2888376
I think the speed of sound is somewhere around mach 1.0

>>2888400
nah, it's around a TNG warp factor of 0.01651

Am I welcome in this thread if I just fuck around with esp32 and don’t have a great grasp of ‘proper’ electronics?

>>2887742
okay an update of what i have been trying so far:
to simulate a switch press im trying to set up a transistor to send a go signal from the signal pad on the button to the chip

i have the base pin of the transistor soldered to a 1k ohm resistor and that then soldered to a wire that is connected to the positive output of my relay

i then tried soldering t he emitter to a split wire that connects to the ground pad of the button and back to the negative output of my power supply

the collector is then soldered to the signal pad but when powering it up it seems to just give a constant signal instead of a single switch press

where have i fucked up?

im at a loss

File: 20250112_113159.jpg (1.98 MB, 3088x3088)

1.98 MB JPG

>>2883138
What is this connector?

File: images.png (7 KB, 354x142)

7 KB PNG

>buy induction cooktop
>has a button to switch between like 300, 500, and 1500 watts
>has a button to change temperatures
>watts button does literally nothing below like 300 F
>temperature button increments in 10F but has obvious dead spots

isn't this just PWM modulation? why are these companies allergic to just doing "0 to 100" in 1% demarcations? why is everything so fucking retarded?

>>2888436
> 300 F
Yes, still using farenheight is retarded, so you can assume it’s the tip of the iceberg.
Mine does 1–10 in single increments (only thing I used), but recently learned 10 isn’t the max… you have to switch it to temp mode, and set it to 450 F for “real max” which I only discovered after 20 years when playing around one day.
Of course, the temp mode has no bearing on the actual temperature—the whole thing is retarded.
Yea, it’s just PWM.

>>2888411
> welcome
There is an /mcg/ thread, but mcu’s don’t do much of anything without talking to the external world which is where people like you generally come and talk to people like us.

File: 1735347241925783.jpg (91 KB, 935x1179)

91 KB JPG

Where do I get reasonably priced electronic water valves that I can proportionally control with a PWM signal?

>>2888420
>where have i fucked up?

your post is one place
to get a meaningful response we'd need 2 things
1) a clear explanation of what you wanna accomplish, and how
2) a diagram (preferably hand drawn) coz your description is useless. for example:
>go signal from the signal pad on the button to the chip
does this signal go to V+ or to ground, and is it steady or temporary?
>a wire that is connected to the positive output of my relay
so, to a contact? normally open? which comes from where? an external supply or the humidifier supply? at what voltage? and what turns on the contact?
>base pin of the transistor
NPN or PNP?
>negative output of my power supply
power supply that drives the relay? or humidifier supply? are the two joined?
>the collector is then soldered to the signal pad
so, the transistor is supposed to drive this ''signal pad'' (whatever that is) to ground?
because wired this way, it cant supply a voltage, only ground a voltage
>a constant signal
''constant'' means nothing. is it ground or V+?
>instead of a single switch press
why would it be otherwise?
if you want a momentary action you have to use a differentiator circuit, i.e. a cap and resistor

File: IMG_7191.jpg (1.38 MB, 3024x4032)

1.38 MB JPG

>>2883138
Does anybody know about decent STM32 tutorials? I have to program a Nucleo board (basically a white arduino from STMicroelectronics, picrel of my lab's haul) in a short amount of time to work as some SoB I was tasked with designing. At this point I've made more application specific Nucleo boards than programs for arduino, let alone for a STM32. I already know how to use the CubeMX to assign pins and all that so I guess that helps a bit, but I genuinely do not know how to write any code my autistic brain basically is only capable to understand analog casuistries

File: file.png (192 KB, 2090x605)

192 KB PNG

>>2888477
i program ST MCUs by hand, never used Cube before. what do you need it to do?

File: Screenshot 2025-01-13 213607.jpg (80 KB, 1081x963)

80 KB JPG

How can I apply a negative offset to a control voltage when I don't have any negative voltages in my circuit? (It's run off a battery)

I understand the concept of a noninverting summing amplifier, and I know I need to add a negative voltage to my input to "subtract" a voltage. But how does that work when my circuit is all between 0 and 9 V? Do I redefine "ground" using a potential divider? That seems like mathematical sleight of hand.

>>2888592
Try a differential amplifier.

>>2888592
>>2888596
adding
>That seems like mathematical sleight of hand
and depending on what you're doing possibly the best way of handling it. What is this voltage supposed to control? are you using a R-R or single supply opamp?

>>2886772
>What's the point of LCSC when the shipping is so expensive?
digikey/other US-based alternatives have shipping even more expensive than LCSC outside of the US.

File: derfffffffffffffffffff.jpg (224 KB, 1023x958)

224 KB JPG

>>2888596
Thanks, that sounds like the right concept, I just didn't know what to search for.

>>2888599
I am this anon >>2885923

I'm still trying to deal with using different JFETs to the out-of-production ones in this circuit. I've breadboarded it all and it works using J113's as replacements, but JFET Q7 still lets through some signal under conditions where there should be no output. This is because Q9 is always passing a slight amount of current, and so the "ARG" signal it outputs is always above 0V. I guess the circuit was designed with the properties of the 2N5457 in mind and the J113 is obviously different in some respects.

So I figure by subtracting some voltage from the ARG signal, I can make the gate voltage of Q7 more negative, and it will turn it off more completely when ARG is not present. If I use an op-amp, I guess I can also use the gain of this differential amplifier to make sure the signal is in the ohmic region of Q7. I've tried some different values for R41, R46 and R47 to try and change the gate-source voltage of Q7, but no success.

I'll be honest anons, I may be in over my head

>>2888607
How is the JFET Q9 being biased in the first place? It looks like with the switch in the Repeat mode R58,63,64 will be what bias the JFET, and when in Manual mode R61,62 are doing the biasing. You'll probably want to do the potentiometer trick again. It looks like the 10k/10k and 150k/150k dividers are making a Vcc/2 bias, so adjusting both of those down a little may solve your problem. Note that changing R63/64 may change the gain set by R58, but it's probably fine.
You can make an LTspice simulation of the ADG circuit to test how changing resistors will influence the output voltage range.

Personally though I'd go back and look for JFETs that better match the Ids/Vgs characteristic curve of the unobtanium component. There's still like 10 commonly made THT JFETs out there.

File: generate negative voltage.jpg (91 KB, 1000x833)

91 KB JPG

>>2888592
>when I don't have any negative voltages in my circuit?

once made a quick circuit similar to this to fix a DVD player whose -12V voltage (for audio) had gone bad

>>2888614
I admit I hadn't considered changing those voltage dividers but you are right, that should solve the issue at the source by changing the output from Q9. I'll give it a go, thanks.
J113 seemed to be a reasonable match in terms of VGS(off) and IDSS. The problem I have with troubleshooting is that all I know about this circuit is it's supposed to make wacky bleep bloop noises

>>2888620
Interesting. The 555 chip sure seems to be versatile

File: SL007DH15B09.png (212 KB, 503x257)

212 KB PNG

I want to repurpose an old screen as a monitor. The screen is an 7' LCD display, 800x480p part number SL007DH15B09. Came from a genesis tablet model GT-7200.
I need the datasheet for the screen to find information on what display control module to buy. Google gives me nothing.
Looking around the original board I found a goodix GT811 touchscreen chip but nothing specific the display part of the screen. It uses a 50 pin ribbon if that matters.

>>2888491
Essentially a HMI. The thing I'm currently trying to make it work is the touchscreen, 4 buttons and a scrolling knob

>>2888607
‘ol ray really loved him some jfet

File: Screenshot_2025-01-14_18-22-14.png (61 KB, 1208x1797)

61 KB PNG

Need some advice bros
Is this an okay way to isolate the 3.7V Li-cell when a USB cable is connected and make the load be powered by the USB cable instead of the cell?

File: 466E0B4E-D18C-4646-91C0-1(...).jpg (78 KB, 1303x587)

78 KB JPG

>>2888713
Usually, I think you can get away with one diode, and two if you’re worried about potentially backfeeding the USB supply (which is probably not a real problem).

In the olden days they had these 3.5 mm power jacks that they re-purposed from audio that had little switches in them that could be used to automatically switch to the plugged-in source.

I assume ridiculous nonsense like the usb-c specifications still indicate you can get 500 mA out of it without writing thousands of lines of firmware code, but I’m not sure.

Is designing a usb hub with dp alt mode and power delivery super complicated, or could a beginner accomplish it, with enough patience of course?
If anyone has an idea for a good starting point, that would also be great.

>>2888713
A li-ion cell will not go above 4.2v (not without blowing up at least). Since that is lower than 5v you can simply place a single diode in series with the li-ion cell. As soon as the usb is connected the diode will become reverse biased and the cell will no longer discharge. You do not need a mosfet. Ideally use a schottky diode to minimise the voltage loss. I've done this before (albeit with a 12v system) so I know it works.

>>2888831
> design a usb hub with PD 3.1 using a usb hub IC
Easy. Buy IC, implement app note.
> design usb hub with PD 3.1 using only discrete transistors
Probably impossible for a single individual over the course of their remaining lifetime

>>2888840
>Probably impossible for a single individual over the course of their remaining lifetime
...
It's not THAT hard anon. Combinatory logic predates transistors and is as old as the first non linear devices. Iirc they were making combinatory logic out of bulbs, even before tubes

>>2888835
But wouldn't i have to contend with the power loss because of the diode drop always

>>2888713
I think that will work. The FET circuit alone only works as a negative polarity protection, but with R7 there it should forcibly turn off the FET when the USB is connected, preventing backflow, and also preventing the voltage drop that you'd get by using a diode on the lithium ion side.

>>2888831
Not trivial. High-speed USB design on a PCB means adhering to differential trace and ground plane best practices to minimise crosstalk and prevent glitches. You'd likely need a 4+ layer board, and 2-sided soldering of small SMD components. Either you buy the stencils with the PCB order, or you get them to solder it for you, both will be pretty expensive, maybe in the $20-60 per board for 5 of them. Cheaper if you can make the stencil yourself. If you want 5 of them, then it might be worth doing if you value your time sufficiently low. If you want 1, fuck no. Just buy one. Watch out for nobrand ones on aliexpress though, it's not unknown for them to contain keyloggers and such. Ones from Ugreen or whoever are probably fine, at least if they're spying on you it's probably at the behest of the CCP, instead of to sell your passwords to hackers.

>>2888852
Assuming you're using a 1n5822 schottky or similiar at a current of 1A (dont know what current you're running at but the smd pmos and USB suggest a low power circuit) the diode will drop ~350mV. If the li-ion cell is fully charged you'll be losing 8.33% of your power on the diode (4.2v * 1A = 4.2W, 350mV * 1A = 350mW divide to get power loss). I'd say the forward voltage drop of the diode would be far more troublesome than the power dissipation but you'd still be fine for a 3.3v system. I'm guessing you're alright with the voltage loss becuase you already have a diode in series with the li-ion cell in your schematic.

Also I noticed that you configured the pmos as a source follower so if you're depending on it to drop the 5v usb supply down to ~4.2v so as not to exceed component voltage ratings then put in a normal silicon diode in series with the usb as well if you're gonna remove the pmos. If your circuit won't be damaged by 5v then you lose nothing by taking the pmos out. The li-ion can't discharge into the usb and the diode prevents the usb from pumping current into the cell.

>>2888866
Disregard the power calculation bit just realised you're already using a schottky with a lower vf than the 1n5822. Also you can verify that the circuit works with the schottky alone using falstad.com/circuit

>>2888866
I think the more important part is the voltage drop. If you're running the lithium ion cell into an ultra-low-dropout regulator like the HT7333 in order to get 3.3V for an MCU or other non-5V-tolerant IC, you'd normally be able to access more than 90% of the cell's capacity. But with that diode drop there, you're only able to access like 60% of the cell's capacity. If you're using chips that will run fine off 5V then my point is moot.

If you're using an audio amplifier or other power output chip, then because power is proportional to voltage squared, you're losing 16% of the maximum total loudness at 4.2V, and 21% when the cell is down to 3.2V.

If an additional MOSFET and two resistors isn't a big deal to add to a circuit, I see no reason not to.

>>2888888
kys

>>2888888
WIN

>>2888888
Sharty won

>>2888888
>>2888891
>>2888892
>>2888893
go back to your kiwifarms ripoff spam bot and pedophile ridden discord circlejerk website with a nonfunctional catalog and minecraft server infested with datamining scripts and groom more minors into self harm while getting datamined by the Taiwanese faggot and the orange pedophile you stupid fucking wojackspamming onions party immigrant

>>2888906
Total.
Baritone.
Meltdown.

>>2888906

testing

if i have a 555 timer in monostable mode set to output a single pulse and connect constant power to it will it still only output a single pulse or will it continually pulse

>>2889005
Depending on how it's wired up, it may make a single pulse when you connect power, or it may make no pulse at all until separately triggered. A monostable 555 should not keep pulsing by itself.

>>2889008
how could i wire it up so it only creates a single pulse when power is turned on?

>>2888840
>>2888864
Thank you for your answers. I had previously been looking to just buy a device but I couldn't seem to find one that clearly stated that it supported all the features. But maybe I'm just dumb.
The thing that most were missing is power delivery out. I can only find hubs that let you connect a power supply to then charge your laptop through the dock. I wanna connect it to a mini pc that has a usb c port with 15W pd and use that to power a ~10W display through the hub. Is that possible?

>>2889005
>>2889010
Actually now that I look at the 555 schematic, I think it will never trigger. I'd do some simulations, but you may need to specifically look for a different type of monostable circuit that can't retrigger when fed a long pulse.
Do you need a delay between power-on and the output pulse?

>>2889012
Is the 10W at 5V? If so it should be supported by basically everything. If it's at 9V or higher, I'm not sure. Maybe browsing chargerlab's products and reviews might show you whether there's certain products or search terms you need to look for. Some gurus on /g/ might know if you're lucky. Otherwise maybe you can take a monitor and mini PC into a shop and test it.

Honestly I doubt you'll find anything that doesn't require an external power input for the hub.

>>2889013
>Do you need a delay between power-on and the output pulse?
it shouldn't need a delay but it probably wouldn't hurt either
thank you for the response by the way its helpful ill try looking for a circuit that can't retrigger

>>2889010
>wire it up so it only creates a single pulse when power is turned on?

to trigger it, you apply a low signal (less than 33% of V+) on pin 2
to do this automatically every time it powers up, you
- put a resistor from V+ to pin 2
- put a capacitor from pin 2 to ground
- the time constant of these 2 components (time = R * C) should be considerably shorter than the time constant of the main timing RC constant (on pins 6 and 7)

>>2889013
I've contacted the pc company, but haven't gotten an answer yet, so I only know that it outputs 15W total.
I'm doing dome diy, so the display is not a full monitor. I was gonna hook up the driver board to the hub, which supposedly needs 5V 3A. So if the output is 5V, you're saying I can hook it up to any of the ports on the hub?

>>2889026
holy based thank you ill give it a try

>>2889028
Yeah if the display wants 5V, there doesn't need to be any fancy power delivery chips inside the hub that negotiate the power given to the display. That said, 3A isn't nothing, I'd want to ensure that the hub is rated for enough output current. Maybe there's a per-port limit, maybe there's just a total hub-wide limit, idk. Conventional hubs may have polyfuses or whatever to prevent the output current going above 2A or whatever.

>>2889030
Yeah this will definitely work, but it will do it as soon as power comes on.

>>2889032
>but it will do it as soon as power comes on.
it should still be fine thank you again very much

ok, another question
i have an analog sensor powered by an automotive vehicle grit and an adc powered by the same grit but its 5v can be switched of. How does one protect a not powered adc from the foreign voltage? afaik max allowed input is VDD+0.3v so if vdd is disconnected and the sensor shats out 1v it should go up in smoke, right?

File: file.png (35 KB, 818x575)

35 KB PNG

for me, its a 431 + pass-transistor.

>>2889045
I think, typically, you’re going to want an opto-isolator. So generally you’re looking at some kind of robust buffer/amplifier that feeds a LED, and another diode picks up the light and you use that as your adc input. There is a literal air gap.
This is because I wouldn’t consider automotive power to be the cleanest of things… you know spark plugs and whatnot.