/diy/ - Do It Yourself

Which electric component do you most identify with?
For me it's the diode.
I just feel it on a spiritual level.

>>1932152
the capacitor
pic related

>>1932154
Based capacitor gondola

>>1932152

i see myself more as a microphone.

>>1932152
imbalanced?

>>1932125
1. How do I find chips I need? I've gone through this process before and it's always super annoying, I used sites to search before but they seem to not be exhaustive.

2. Specifically I am looking for a chip that does the inverse of this https://www.ti.com/lit/ds/symlink/cd74hc154.pdf which is pic related. Basically takes in the decade counter's input and turns it into binary.

>>1932189
Is that a ring counter? Why not use counter with outputs already in BCD format and use a 74154 to get it into the 1-of-10 format if needed. Otherwise I think what you want is some sort of multiplexer.

>>1932194
the 1-10 format is the output of a keyboard matrix.

Actually while I'm typing this you gave me an idea so that I can avoid using a new chip so thanks for that.

>>1932198
Oh you want to encode a single keypress into a BCD value, presumably with some extra way of telling if any buttons have been pushed at all. It's a 3x3 then that number 10 could used for that.

Actually what you want is called a "priority encoder", and the 74147 10-line to 4-line priority encoder fits the bill there, I think.
What's the new idea?

Another option is to just use a multiplexer like a 74150 and count through all the states synchronously with a BCD counter, or to do a similar thing with a couple of 74165 PISO shift registers. But for asynchronous methods then something like a 74147 is probably as good as you're going to get. Might be decent selection in the 4000 series too.

Just go to the wikipedia page for "list of 7400 series integrated circuits" and ctrl-F your way around. One also exists for the 4000 series.

>>1932152
i am most definitely a germanium transistor, slow, leaky as fuck and outdated

>>1932165
that face has seen some shit

>>1932152
A bipolar tranny

Brainlet here, have two questions: first one, if i have a power source like a solar panel or a fan or whatever and i know for a fact that they do work and are producing some energy, how can i calculate their maximum power output? For voltage i think i can just measure it straight away but what about current? If i just put a random load on it it will skew the measurements, won't it?
Second question is the opposite of that: is it possible to figure out how many volts and amps a load needs in order for it to operate properly if it has no stickers or a manual or something like that? For example, if i have an old radio, how can i know if it works on 110V or 220V if there's no sticker with that info on it?

>>1932324
>first question
usually for PV systems you just short circuit them. (to get the current) and measure the open terminal voltage for some condition to get the voltage. THose two are also important measurements in transformers.

>>1932152
coin cell that leaked and corroded itself

>>1932341
Isc and Voc are some of the specs that are used to describe the panels, but they don't tell you the maximum power.
>>1932324
The answer is neither of those are easy.
>If i just put a random load on it it will skew the measurements, won't it?
Yes. You can get a MPPT controller which will automatically adjust the load and try to get the most power out. They often have a display that will tell you the current power output. Of course that still doesn't account for things like pointing the panel in the right direction and how bright the sun is on a particular day.
>if i have an old radio, how can i know if it works on 110V or 220V if there's no sticker with that info on it?
plug it in and see if it catches fire

>>1932324
>if i have a power source like a solar panel or a fan or whatever

your first question will highly vary depending on the source provided...a solar panel has a much more complex or just plainly different power curve compared to a fan...also it depends much on the type of load is connected to it

for most uses if you have ean absolutely radnom source that you know nothing about it will be sufficent to just measure voltage and short circuit current(if the source allows it safely) or connecting a simple resistor load and measuring what happens to give you an approximate or close to approximate power rating

also btw most sources like transformers and electric motors/generators and different kinds of power sources have standardised tests for checking their characteristics which most of the time are realtively simple and require only a DMM, a pencil/paper and some time

> if i have an old radio, how can i know if it works on 110V or 220V if there's no sticker with that info on
the safest way of figuring out those kinds of things is to open it up and check what components are inside....a lot of the time just by looking at capacitor values is enough to gauge it , also checking the specs of diode bridges, checking transformer windings , transistors etc

otherwise you need to get a variable transformer and just slowly raise the voltage untill it works for AC and a variable lab power supply for DC and just raise voltage/current untill it works

I know there are 555 and 775 DC motors, but where do the names come from? And are there more?

>>1932428
>but where do the names come from?
from the manufacturer
>And are there more?
more what ? types? in stock? motors in general?

>>1932433
>from the manufacturer
There are multiple.
>more what ?
You really didn't understand my question at all, did you? I'm asking about de facto standardized motor dimensions.

>>1932436
i understood your question but it was generally vague and both could be easily anwsered if googled but its infinetly more amusing to poke back instead of spoonfeeding you google results

>>1932152
an opamp with floating inputs and no negative feedback

Is there any disadvantage to stacking multiple op-amps for large gain? On the face of it, it seems like some op amps have a poor frequency response if gain is (e.g.) 6. Why not have three op amps each tuned to a gain of two?

The only bad part of this I can think of is some sort of input delay for each op-amp, meaning the max frequency you can obtain tanks. Am I completely wrong?

>>1932472
generally op amps cna be stacked and are stacked
the downside is the input delay you said and the amplicifation of noise in each stage

>>1932472
Yes, op-amps are defined by a gain-bandwidth product, so the open-loop-gain will decrease as a function of increasing frequency. In the case that you need more gain that your op-amp will allow at that frequency, or just want the PSRR that comes with a lower gain, then it may well be worth chaining op-amps. But I'd look into using a faster op-amp before that.

Another reason is for input impedance reasons, such as in the example of an instrumentation amplifier.

Thirdly, sometimes you need to combine op-amps with different properties, say a high-precision op-amp that can't source much current alongside a lower-precision op-amp that can source much more. Or maybe a JFET op-amp for extremely high input impedance as a buffer/filter chained with a higher frequency op-amp for more stable gain.

>>1932472
Uh yeah all the time. Not only opamps but any sort of amplifier. Gain does not add, it multiplies so sometimes it is a much better choice to use 10*100 to get 1000 than making something with a 1000 gain out of the box. (1000 is a small gain, but you get the idea.) Also
>impedance matching

I have a small diesel powered generator that produces 42 volts AC at about 2kW. It was used for street lighting or something special. There's 5 wires coming out of the generator. Two thick ones and three thin ones. The thick ones are connected to the load and also connected to a big AC capacitor. The three thin ones are all connected to eachother by two diodes and nothing else. The generator has two slip rings inside.

I want to rectify the output and use it to charge 24V batteries. Therefore i want to lower the excitation of the rotor coil so it produces less voltage in the stator coil.

But what kind of generator is this? What's the function of the two diodes and the capacitor?

>>1932523
the capacitor is used for starting the motor...it causes the current going into the windings to lag which in turn helps start a rotating magnetic field which causes the rotation of the rotor

>The three thin ones are all connected to eachother by two diodes and nothing else
can you post a picture ? they might be control signals if the motor has a controll box on it or could be some kind of signal outputs for voltage or an on lamp

>Therefore i want to lower the excitation of the rotor coil so it produces less voltage in the stator coil.
just rectify the output and connect it to a buck converter or voltage regulator ..if you are gonna charge batteries you will need some charging circuit anyways

>>1932532
>start a rotating magnetic field which causes the rotation of the rotor
It's an alternator, not a motor. It doesn't need help rotating. Diesel will do that.

The three wires and diodes are not used for signaling. Without these, there's no power.

>>1932539
What's the manual say

>>1932539
fuck me im either blind or retarded i didnt even register the fact its a generator

>>1932549
Nothing because there is none. There's not even a product name on the alternator. It was a small production batch of custom power generators probably.

>>1932555
Timer trips detected

>>1932555

>Timer trips detected
probably my biggest achievment honestly

Hey, i need some help finding the values for these caps. Ive tried googling just whats writted on them, "smd cap markings" anf all that and I havrnt been able to find anything.

>>1932555
I'm more familiar with motors than generators myself too. I do understand induction motors and permanent magnet generators. But induction generators is something different. You must somehow build up a magnetic field from nothing but rotating copper coils.

>>1932556
Then draw up a not-retarded schematic so we can look at the circuit. Hard to know with just what you've said.

The three wires + diodes are probably rectifying the voltage to the generator capacitor. The generator capacitor is managing the phase / maintaining the induced current.

>>1932563
big ones are 220uF 4V
the right one is 3.3uF 50V
the left ones are 47uF 4V

the markings are capacitance (the numbers that are in the standardised series )
the other number is voltage rating and the letter is the batch series/electrolythic type/usage series type (basically unimportant unless you need specific capacitors)

>>1932572
>You must somehow build up a magnetic field from nothing but rotating copper coils.
you dont
the cpacitor is still there to start the motor/generator
the small wires are DC (the diodes)and they are connected to the sliprings that power up the rotor which starts a magnetic field (the engine rotates the rotor) and induces a voltage into the stator windings

>>1932581
It must be something like this then.

Still not sure how a capacitor can start a magnetic field in the stator if it's uncharged.

>>1932154
I really like this image>>1932532
The diodes and capacitor might be some sort of voltage regulation going to stator coils? Like a car’s alternator. Doesn’t really matter whether the rotor or the stator is doing the permanent field, IIRC.

Could also be for an induction generator, maybe.

>>1932633
Either way, if it doesn’t have permanent magnets in it it will need some way of creating a magnetic field, probably through a battery. Might be something like a centrifugal switch so it doesn’t drain when the motor is off.

RIP that formatting

>>1932125
Virgin 6 pulse inverter. Vs CHAD 12 pulse SCR

>>1932637
I imagine the body of the motor must have to be magnetized a little bit to start with. It wouldn't need to be much, just more than zero to get it started until the current is flowing.

>>1932637
>>1932687
The diesel engine connected to it has a battery for starting you smooth brain of course its not pulled from thin air

>>1932700
OP here. No it's started up old fashioned with a hand crank you lazy mofo. No batteries.

>>1932713
Then theres a charge carried in the capacitor ..you dont need a lot to start up a rotating magnetic field since it builds upon itself

How do I into light switches? I added on a room to my house and I have no idea what I'm doing. Pic related. I can figure out from tutorials how to wire one of the switches, but I want both on the same circuit.

>>1932720
google parallel switches

>>1932718
Yeah but that charge will leak through the coil it’s connected to unless there’s some kind of centrifugal switch or starting button in place. Even if there is, the capacitor charge will still leak over a few days or weeks.

But I’m not sure how diesel engines keep their glow plugs hot if they don’t have a battery, maybe there is electricity somewhere.

>>1932713
What year is that generator from..i never saw a diesel generator started by a hand above 500W...you are missing something then

>>1932758
It’s 40V, makes me wonder what kind of current it drives.
Based SI units for power.

>>1932758
It's a Hatz diesel from Germany. Will post a pic later (its dark now).

>>1932762
About 47 Amps.

>>1932763
>About 47 Amps.
fuakk, they shoulda used a higher voltage
those wires are gonna be hella thick
>Germany
Looks like East Germany, which would explain why such an old-fashioned piece of kit is still around. I think I'd quite like to go to what was East Germany, sounds like an interesting combination of communist economic restrictions and german autism.

So I bought this really cheap hdmi-to-av converter. It works, but after a while it gets too hot and it'll just show colorbars for a second or two before coming back on. I took the thing apart and used a thermometer on it to see exactly what was getting hot. There's a couple tiny chips (like 2.5mm) on the board that seem to be getting the hottest. How would you go about cooling these? Do they even make heatsinks that tiny? Should I just cut a hole in the enclosure and glue a small fan on top?

>>1932822
I forgot to add the image

>>1932822
you're better off getting one that doesn't get hot. That indicates a serious design flaw.
You don't need a tiny headsink, you just need to be able to mount any heatsink.
And yes, they do make small heatsinks. But that won't necessarily solve the underlying problem.

>>1932822
Show a picture of the components. If they're resistors or inductors or something shitty like that then you can desolder them and replace with higher power resistors. That or stick a heat-sink atop them.

>>1932822
>>1932824
Ain't no fucking way one or two tiny 2.5mm chips will make that significantly larger case feel even remotely warm, even if said chips are running at 200F they are just too small to warm up their surroundings to any significant degree. Some larger component has to be getting hot. A TO-220 package or at least a SOT-223 or something

>>1932850
The two things highlighted with red seem to be getting the hottest

>>1932581
no one in their right minds makes single phase induction generators and they must be grid connected because of something that caused a lot of debate in the last thread
>they always draw reactive power from the grid

>>1932853
Oh, those are ICs. I thought you meant one of those two-pin little guys. Definitely possible to put a heat-sink on one, but maybe a little forced airflow would be easier.

If you can read off what's on the two ICs that would be interesting, but not particularly necessary. Only in the odd chance that one of those isn't meant to be getting hot at all can we determine anything. But considering it's a video converter running at 1080p, it's going to be clocking pretty fast and going through a fair bit of data, so some heat is expected.

>>1932727
Okay. What you're describing is how to wire two switches to operate light 1. I know how to do that. What I want is to add light 2 further down the circuit from light 1.

>>1932125
I need a book with everything there is to know about DC and AC, and TONS and TONS of circuits to solve, i really need to master this concepts asap. Is there any book like this out there? Or youtube channel?

>>1932900
https://www.amazon.fr/Nilsson-Electric-Circuits_10-James-W/dp/0133760030

>>1932905
>>1932900
Actually this one is a little bit less worse
https://www.amazon.com/Fundamentals-Electric-Circuits-Charles-Alexander/dp/0073380571

>>1932905
>>1932906

cant vouch for either but both are available on the pirate bay.

>>1932910

crap, they're both full of this shit which is academic masturbation, and has zero application in real life.

>>1932900
read the OP?

>>1932900
>I want to learn everything !!! even the most complicated stuff!!!
>>1932915
>ewww not that i dont want to learn the basics and understand the science behind it i just want to get knowledge!!!!

kill yourself unironically

>>1932920
Well it's good to know how to tackle a problem like that. But if the book is full of that kind of abstract problem, and at the expense of examples that are actually useful, then it certainly is a reason to avoid it.

Is "full of this shit" a hyperbole?

>>1932923
>>1932900
>>1932915
90% of the exercises in circuits and electronic books are for you to try to get a "muscle memory" for the rest of stuff. You asked for that I gave you that. The purpose of exercises are to train. If you can't do 70% of them then you better find something else to do.

>>1932931
And if you show me one of those weird exercises I will probably huff and puff, take a while but when my intuition (built by actually doing the damn thing) kicks in I will solve it in the end.

>>1932923
>and at the expense of examples that are actually useful,

there is no textbook that teaches you "useful practical things" at least not like people insist on doing

pick an active electrical component and get its datasheet and copy the "usual use case schematic" ...congrats you made a cricuit

otherwise learn electronics on a fundamental level and do your own shit from the laws and effects you learned...its as much of an art as it is a science which means if you have no imagination you will never be good at it , you will just be an equivalent of a "script monkey" , just copy pasting other designs because they work not because you know how they work

>t. engineer that learned how to do it by applying the "useless science work" by myself in real life circuits

>>1932523
Found something myself which may be the answer. This kind of motor is not very common i guess because it's hard to find anything.

https://www.worldpowerfaqs.com/en/categories/alternator/123-how-does-a-capacitor-assisted-alternator-work

>>1932936
I might have lost track of the replies but you ever said what KIND of generator you are using?

>>1932938
I don't know what kind it is so i couldn't search for the workings of it.

>>1932941
Do you have it with you? Pictures or anything? It MUST have at least a plate with the power rating and some information

>>1932936
>>1932941
yes thats the generator ...

i cant believe i forgot about remanent magnetic rotors ....looks like i need to dust off my electrical machines textbook and brush up on it

I am working on repairing an old record player from the 50's and I am replacing all the old paper capacitors with new film capacitors. There's this one here on the secondary of the output transformer that I am struggling to find an exact replacement for. Does anyone know what it is for and how critical the value of it is? I'm thinking that it's some sort of filter, but I'm not really sure to be honest; I've never seen a capacitor like this on the secondary of an output transformer. I'm probably just gonna go through the .033uf and .047uf capacitors and find the closest one to .040uf to use.

>>1933036
its a decoupling capacitor to filter high frequency induced shit ....drop in a 100nF or 10nF ceramic capacitor and it will be golden

Why do people say K620 instead of 620R? Is the world just broken like that?

>>1933043
>Why do people say K620
where ? ..i never heard of that in europe

>>1932853
HDMI is a digital-to-analog conversion, and the larger of the two IC chips is right by the video output (and country selection switch) so its probably some analog driver for the video.
As the other anon said, its most likely a poor design.
Yes you can put a heatsink on it, but depending on how much heat its putting out, a small fan directly over the chips should keep them cool enough to run well. A simple 5v 50mm-ish fan (something for a raspberry pi) will be perfect.

>>1933102
I meant to start that post with "HDMI to composite is a digital-to-analog conversion"

It is literally impossible to get binary or serial input into a modern computer without using an Arduino or Raspberry Pi. FACT. if you disagree you are a low test beta NIGGER

>>1933043
Never see it like that before. K is usually for relays/contacts (switches), R for resistors obviously.
I've seen it used in place of a decimal in resistor values, so a 4.7K would be typed at 4K7. It was done that way for a long time because back when photocopying was still crappy, small decimals may not copy, OR some speck of dust could be copied and look like a decimal on accident.
So they started using the K as the decimal to help with readability. So maybe K620 means .620k ohm.

>>1933104
>binary or serial input
Serial is binary
2/10 made me reply

>>1933112
>what are parallel ports

binary=1 bit
serial=bitstream

>>1933118
still binary

>>1933119
>acceleration is a unit of distance
>wattage is a measure of voltage
>bytes are a measure of resistance

>>1933120
you're not making any sense, whatever point you are trying to make isn't being made.
All digital data is binary. See what happens when you list arbitrary facts without proper context? They don't mean anything. They are just true.
Serial data is binary. As is parallel data. Without more context, you are just making retarded posts.

>>1933120

>>1933124
>words are made of letters. therefore, words and letters are the same thing.

>>1933131
you'[re just making my point. You don't seem to understand enough to grasp the different contexts.

>>1933131
It's more like latin words are made of latin letters and thus are latin
cyrillic words are made of cyrillic letters and thus are commie bastards

>>1933124
>All digital data is binary.
back to school please
digital is discreet, binary is a condition of digital where there are two discreet levels.
there is nothing to say that either digital or a serial stream has to be binary specifically.
if you want to argue that a serial stream of data has to be digital i would disagree with you but i wouldn't care enough to argue over it.

>>1933164
either way, the obvious context here is a delineation between devices that are either on or off (like a switch) versus things that provide byte-level data (modern sensors with encoders)

>>1933164
1/10 for being technically correct. 0/10 for misspelling discrete.

Imagine being this triggered and baited over such a blatantly troll and, by now, overused meme post

>>1933172
Contextually speaking, you could differentiate a parallel data stream where each line is binary, and one where each line is trinary or analog or whatever. I think it would be pretty easy to agree that modern computers are binary in nature, in comparison with analog computers.

Is a decatron tube binary or decimal?
An SP3T switch? Obviously it’s a matter of semantics with no bearing on reality, so I think the takeaway is to neither call parallel “non-binary”, nor to explicitly call someone out for doing so.

Oh but of course, binary doesn’t just mean “0 or 1”, but rather it’s a base-2 numerical system. Convert decimal 9 into binary and you get 101, so a parallel line coming out of an ADC or whatever could definitely be said to be carrying an N-bit binary number.

No clue what parallel port pin outs are actually like though.

Hey guys, sorry if I'm asking this in the wrong place. I'm looking to learn digital design\logic design, is there any textbook recommendations? I genuinely want to learn this but the textbook that my class use is so absolutely fucked that it becomes a headache to understand. I noticed that the principles section of the OP didn't have any for this, again, sorry if I'm not in the right place.

>>1933312
I'd recommend digital design and computer architecture by harris and harris.
It's on libgen if you just want to pirate it, but I bought it since I like it so much.
I think there's also a solution manual on libgen too.

>>1933317
Got it, thank you so much. Apparently, there are two different versions. The MIPS (the one you have) and the ARMS (pic attached) version, any idea what's the difference or which one to pick?

>>1933312
For lower level stuff, “Digital Fundamentals” is what my digital logic physics paper followed. Didn’t go near computer architecture, but it did touch on karnaugh maps, shift registers, master/slave and synchronous flip-flops, that kinda stuff.

>>1933318
I think the difference is just for the latter third which focuses on computer architecture.
I read the mips one, but I'd assume the arm one is good too.
Mips is definitely more straightforward than arm, so that might be your best bet if you want to read that section. Otherwise if you just want to focus on digital design then both will work.
Also, look up the program "logisim-evolution". It's a logic circuit simulator that works well in my experience.

>>1933329
>>1933317
My indecisiveness is going to kick in at this rate, both books are much better than my horrible college text. I'll just read both and cross-reference my learning. Thank you both so much, I was about to go under a breakdown when my professor went from decoders to gate/logic array in two seconds.
>>1933339
Okay! I already have logisim downloaded so the transition shouldn't be too hard. Thank you two again.

>>1933343
lol are you the kmap-guy from 2 weeks ago by any chance

OP here, should I put any of those digital textbooks in my paste for next thread? Those digital ones are more specific than the existing textbooks we have, so maybe there should be a section for more specific texts, one for each subfield.
The current OP length is ~1500 or so, there's definitely room for more.

>>1933360
Can't see it being harmful. Probably belongs more in /mcg/ tho.

>>1933206
>where each line is trinary or analog or whatever.
I see you've never heard of a serial port.

>>1933208
>Convert decimal 9 into binary and you get 101, so
I see you've never heard of a serial port.

>>1933104
>what is a parallel port
17 pins of semi-flexible GPIO is what
With significant downsides, one being that modern computers don't have parallel ports.
also
>what is a USB UART

>>1933390
>I see you've never heard of a serial port.
I was going to post something about RS232, but I couldn't remember whether it just left the line at whatever ±12V would signify the end of the end-bit or if actually returned to 0V.
Last I remember, measuring an unused RS232 output gave me something other than 0V, so I decided not to post it. But maybe that was just the CTS/RTS signal (whichever is the output) being constantly on.

Either way, the 0V signal doesn't really carry any data that isn't conveyed by the start and end bits, so I wouldn't call it true trinary at the very least.

>>1932436
Maybe you want NEMA motor sizes? jewgle that

>>1933415

there's no 0V on the data or control lines ever. only on the common ground pin. all signals are always +3 to +25 or -3 to -25.

the end bit is high, a start bit is always low. so a full byte sent is
- 1 low start bit
- 8 data bits (7 in olden days)
- 1 or 1.5 or 2 high end bits

>>1933415

there's no 0V on the data or control lines ever. only on the common ground pin. all signals are always +3 to +25 or -3 to -25.

the end bit is high, and it stays high (idle) until a new byte is sent; a start bit is always low. so a full byte sent is
- 1 low start bit
- 8 data bits (7 in olden days)
- 1 or 1.5 or 2 high end bits

This is a stupid but honest question.

Regarding ATX power supplies for desktop computers. What makes a power supply capable of providing a particular amount of power?

For example, lets use the Corsair CX series, a budget, but common PSU. It's available in 430, 500, and 600 watt models. What exactly is the difference between these other than the available connectors? Which specific components inside allow the 600w model to provide more power power than the 430w model? Is it just arbitrary, and could I somehow "unlock" the 430w model to allow 600w?

This is a stupid but honest question.

Regarding ATX power supplies for desktop computers. What makes a power supply capable of providing a particular amount of power?

For example, lets use the Corsair CX series, a budget, but common PSU. It's available in 430, 500, and 600 watt models. What exactly is the difference between these other than the available connectors? Which specific components inside allow the 600w model to provide more power than the 430w model? Is it just arbitrarily limited, and could I somehow "unlock" the 430w model to allow 600w?

>>1933459

silly question. if they could produce a 600W power supply at the same cost as a 430W supply, then they'd mark all supplies at 600W and kill the competition.

so no, you cant unlock extra watts coz it needs more expensive, bigger, components and heat-sinking.

>>1933460
ok, but why? What part's are different between them?

>>1933457
Oh, just made myself look like an idiot then.
What was he talking about when he said "I see you've never heard of a serial port" then? The fact that it has more than one line that technically sends information per direction?

wait 101 is 5, not 9

>>1933459
>>1933461
The main power limiting parts will be the flyback transformers (big and relatively expensive, especially if they need to be toroidals), and the switching MOSFETs. The higher power models will have larger inductors (thicker wire and more area for dissipating heat, but also maybe larger size and/or different core to avoid magnetic saturation), and more and/or higher-spec MOSFETs. Or maybe just better cooling, as would be the case if they put better heat sinks or heat pipes or fans or whatever. Higher power PSUs would also need larger input capacitors and maybe more mains EMI filtering.

Arguably you could "upgrade" one to run at higher powers by adding better heat-sinking, maybe replacing some of the MOSFETs, fans, etc. but it would be a pretty dodgy refit unless you knew what you're doing. If you had access to the spec sheets of all the parts used in the PSU and knew the schematic too, then you could possibly upgrade it for cheaper than buying a different one, at least to some extent.

It's also possible that the switching controllers are different, which may well be the case if different switching profiles are preferred for higher powers.

>>1933459
Usually the transformer, diodes and filters. (so the whole fucking thing). I've designed a Flyback supply once (100W only) and it was a iterative process, specially choosing the transformer core, snubbers and switching elements.

I cobbled together a 4S battery pack for my 10y/o cordless drill as the original NiCd pack died long ago. Turns out the dropout on the batteries when I load the drill enough (or throw it to full power from 0) is too much for the chinesium BMS and it shuts off. How can I get around this without getting 18650s rated for high drain? The drill starts without issues if I bypass the BMS, but I don't want to open up the pack every time I want to charge it. Sure, a charging port per battery is an option, but also a pain in the ass to add.

How do I go about fooling an inductive loop detector that a car is standing on top of it?

>>1933482

you sound like the guy above wanting to unlock the hidden capacity from your batteries. maybe the BMS knows what it's doing, and shutting down is the appropriate thing to do if you dont want fire and smoke.

learn to live with a weak drill coz of weak batteries. slo your drill, bitch.

>>1933482

you sound like the guy above wanting to unlock the hidden capacity from your batteries. maybe the BMS knows what it's doing, and shutting down is the appropriate thing to do if you dont want fire and smoke.

learn to live with a weak drill coz of weak batteries: slow down and make it home alive.

>>1933486
Get a sheet of metal and put it above it ...there is not particular "fooling" it ..its just a ground loop connected to a preamplifier thats been fed to a microcontroller that opens and closes a relay switch ..some can be adjusted for sensitivity but otherwise you just need a big enough sheet or piece of metal...only problem is you might need multiple if you have more loops

>t. I make electronics for it

>>1933459
Its a fair enouh question i suppose if im guessing you are a PC guy that has no background in electronics and is surrounded by tue overclocking tradition of that community

Basically there is nothing to be overclocled or juiced out of a power supply and you have no business opening it up in your case...they are dangerous and deadly even when not connected because the output capacitors store enough energy to kill 5 people connected together by holding hands

To be able to source more power the supply needs to have its main components upgraded like the transformer , diode bridge , capacitors , output transistors and most likely the controller...which is pretty much everything (thats not considering a PCB witht thiccer traces and interconnect wires and bigger fans and bigger heatsinks)

And the final point is even if you knew what you are doing and were able to do it effectively you would most likely pay more for components than you would for a new supply

Also considering the power supply is the most critical part of the PC stability and overvoltage and overcurrent protection thats the last thing i would go around nigger-rigging

>>1933486
my van was too tall to trigger some of those, just get lift kit for your car

>>1933486
Ive heard of people gluing the strong rare earth magnets under their small cars / motorcycles to help trigger it. No idea if it works, just heard about it.

>>1933521
I get that, but wielding a large sheet of metal is impractical. Someone made a device that seems to do what I want, but it looks to be vaporware: https://www.veloloop.com/

Any idea how it would work?

>>1933350
Uh oh you caught me...
I think it was me? Not sure, if it's related to logisim it was probably me! Anon you have very powerful memory.

>>1933482
Either your BMS is undersized for the battery pack, or the battery pack is undersized for the drill.
Figure out which it is, one or both of those things need an upgrade.

>>1933618
The BMS is 40A. I scoped out the voltage on the pack and it's most likely the cells being undersized. They appear to have quite a high internal impedance for this application. The pack drops a whole 10V when I throw the drill to full power and that's what makes the BMS cut power.

I think my bench power supply's ground terminal might be fucked. Is there a simple circuit that incorporates positive, negative and ground so that I can test it?

>>1933736
found out what the problem was. Turns out I was using a single rail bench power supply when I really wanted to use a dual rail. It seems that single rail would do 0 to 20V while a dual rail would do -20 to +20V. Is there a benefit to single rail over dual rail or is dual rail just a straight up improvement?

>>1933768
straight up improvement. It's extra parts so extra cost.

>>1933486
Maybe something like an e-bow? I.e. receive the AC signal, amplify it, and transmit it again in-phase?

I read a manufacturer's guide which says a common problem is to not provide a return pathway for the DC component of voltage when using a coupling capacitor.

If I have a pin (Max_Out) and don't want to bias the output, should I use a resistor to ground or leave the input as only the coupling capacitor?

>>1933808
Depends on how the input acts. You might find that the input’s own biasing slowly pulls the line too high low for linearity. You could test this by connecting such an op-amp up with no signal up to power and seeing how the voltage floats, but note that your probes will pull the voltage too.

What you’ve made with a DC blocking capacitor is also a high-pass filter, whose corner frequency is a function of both the op-amp input impedance and the capacitance. Too high an input impedance could lead to that corner frequency being too and letting through low-frequencies that you’d rather block, and its response time after turning on the device could be particularly slow.

For both reasons, I’d put a large value (100k or so) resistor to the GND rail. Op-amp biasing should be able to be adjusted around that resistor if it’s an issue.

>>1933808
Wait you’re talking about output biasing? Effectively no bias current is going to leak back through the capacitor, unless you’re talking about putting a resistor before the capacitor. So long as it’s a push-pull output, it shouldn’t require such a resistor before the cap.

>>1933609
i never heard of that but if it manages to make a strong enough magnetic field then yes absolutely i see no reason why it wouldnt

>>1933486
strong magnet

>>1932189
https://www.onsemi.com/pub/Collateral/MC14028B-D.PDF

>>1932189
I think you have to use 74HC154 or two 4051 and use a ripple counter such as 4040 to scan through all the possibilities and stops when the proper input is found.
Let's assume that all 16 possibilities are low, the 4040 scans through all 16 possibilities without finding one that is high so the counting continues. When one of the 16 inputs is pulled low then when the counter input connects that low input to the output of the multiplexer, that stops the counter at the proper binary count and that low signal also signals that the circuit has a valid decode output. Follow that?

Old Atari keyboards were scanned similarly but had 64 possibilities. The keyboard was scanned by two 4051 with the abc of the 4051 sequentially counted through all 64 combinations by the computer, when a closed switch (depressed key) was found, the counting was stopped and the common analog output from one of the 4051 signaled the system that a pressed key has been found.

>>1933808
Your amplifier needs to be something like this:

>>1933987
correction:

When one of the 16 inputs is pulled high then when the counter input connects that high input to the output of the multiplexer, that stops the counter at the proper binary count and that high signal also signals that the circuit has a valid decode output.

>>1933987
>I think you have to use 74HC154 or two 4051 and use a ripple counter such as 4040 to scan through all the possibilities and stops when the proper input is found
or he can just use a 74HC147 or some other priority encoder

if he's doing synchronous, I prefer the PISO shift register route to the counter + demux route, but that's just me.

>>1934025
>or he can just use a 74HC147 or some other priority encoder

I'm sure there's a way to go with 74HC147, didn't know about these but I remember ICs called priority encoders so that deserves a look (never used one).

>>1934073
I found this, looks like what he's looking for.

>>1934073
>>1934076
The 74HC147 is exactly what he's looking for. Almost.
http://pdf.datasheetcatalog.com/datasheet/philips/74HC_HCT147_CNV_2.pdf
Even though it says "10 bit", it only has 9 inputs. All 9 of those being high (active low) causes the output to show binary value 15, a jump up from 9. Which would be a bit of a pain with a 10-key keypad, but I think it's just 1-3 simple gates to add another digit, assuming priority isn't a concern (it isn't). Possibly easier than those three ORs.

On the other hand, the 74HC148 8-to-3 priority encoder has the full 8 inputs, and an extra pin to signify when none are active. Be nice if the 147 was like that too.

On the other, other hand, the 4000 series version of the 74HC147, the CD40147, does do that properly. So it is exactly what he's looking for.
https://www.ti.com/lit/ds/symlink/cd40147b.pdf
So anon, you need a 40147.

>>1933036
Looks like a coupling capacitor for the feedback circuit. I see an 18k resistor that goes off the image at the top that probably connects to the cathode of an early audio amplifier stage.
Larger (.047) should be OK but if you can dig up a .039 or parallel a couple that get very close such as 0.033 + 0.0068 then go for that.

>>1934092
Seems to me that the only one that expands properly to 16 inputs is the MC14532, I'm not seeing how to get the others to do this properly.
Problem with the 14532 is SOIC package only.

I can see the PISO shift register method and the 4051 scanner method but not 40147.

I'll see how my head works on that in the morning..

>>1934108
16 inputs? He only wanted 10 IIRC.

>>1934108
>>1934110

See: >>1932198
>the 1-10 format is the output of a keyboard matrix

>>1934111
OK .. I got caught up with his 1-of-16 logic chart and missed a the post about the keyboard matrix above.
All good .. I still learned something tonight.

>>1933711
Alright I tried a functional laptop pack. 3S 2P works very well with no issues.

Coax and other transmission lines have a fixed impedance for all frequencies because of wave behaviour and all. But what does the impedance in speakers actually mean? Is it 8 ohms for the whole operating range?

>>1934245

yes, exactly 8.00 ohms at all frequencies from zero to infinity.

>>1934245

yes, exactly 8.00 ohms at all frequencies from zero to infinity and beyond.

>>1934253
but your picture shows the oposite of what you said

Hey, so I've been looking at making some of my own SCART cables and I have most of the knowhow skill, but one thing confused me on some of the pinouts I've seen. It says on this diagram that the luma line has to terminate in a 75 ohm resistor, then ground? It says that multiple lines have to terminate in a ground, is that a common ground between them? Thanks.

>>1934245
Note that the characteristic impedance (e.g. 50Ω or 75Ω) isn't the actual impedance of a piece of coax. It's the impedance of an infinite length of coax, or the impedance of a finite length of coax connected to a matched terminator.

>>1934266
For that cable, pins 5/6 on the SNES are the common ground, so the cable is 8 cores plus screen. At the SCART end, the screen connects to the plug screen, all the ground pins, and the resistors (i.e. anything marked with a ground symbol on the right of the diagram).

For SCART-to-SCART, the various grounds should be separated. In a decent cable, each signal gets its own coax, so e.g. the blue signal is a coax with the centre connected to pin 7 and the screen connected to pin 5, and the outer screen on the plug is connected to an outer screen on the overall cable. Cheap cables just connect all the ground pins and the plug screen to the cable screen.

Similarly, SCART to multiple BNC/RCA connectors should use multiple coax cables with each ground pin on the SCART end connected to the corresponding screen.

>>1934284
Ah I see, that makes more sense. I have plenty of coax cables I can use for each video and audio channel. The problem I've had in the past is with cheaper cables producing a buzz through audio when voltage of the video lines is high e.g. a white image. I also have quite a few old SCART-SCART cables lying around that are very thick/well shielded, would chopping these up make for an easier time do you think?

https://en.wikipedia.org/wiki/44,100_Hz
>The Nyquist–Shannon sampling theorem says the sampling frequency must be greater than twice the maximum frequency one wishes to reproduce. Since human hearing range is roughly 20 Hz to 20,000 Hz, the sampling rate had to be greater than 40 kHz.
>In addition, signals must be low-pass filtered before sampling to avoid aliasing. While an ideal low-pass filter would perfectly pass frequencies below 20 kHz (without attenuating them) and perfectly cut off frequencies above 20 kHz, such an ideal filter is theoretically and practically impossible to implement as it is noncausal, so in practice a transition band is necessary, where frequencies are partly attenuated. The wider this transition band is, the easier and more economical it is to make an anti-aliasing filter. The 44.1 kHz sampling frequency allows for a 2.05 kHz transition band.

What sort of filter does 60dB attenuation across like a tenth of a frequency decade? Surely not an analog one?

>>1934308
huh, why not. you mean a notch filter? analog are always superior to digital.

>>1934245
impedance is usually measured at 1khz if not stated otherwise

>>1934308
i still dont understand your question after reading it for 50 times

there is a lot you can do with capacitors and inductors if yo ustage them well enough ..it just depends on waht frequencies you want to attenuate and what kind of filter you need

>>1934308
uh, you do realize that the faster the attenuation the smaller the transition band right?

Do you know any resources on where we can read on how circuit simulations are made? I am trying to do something that gets the DC operating point but I am getting stuck at the basics.
>how to represent circuit as a matrix
>how to linear eqs out of that

>>1934361
yes, I'm wondering what sort of low pass filter is capable of a transition band slope of like -500dB/decade or even more, which seems necessary to get sufficient attenuation if the corner frequency is at 20kHz and the transition band is only 2kHz wide

>>1934342
From what I've read, it's extremely difficult to realize analog filters higher than 8th order, that is, with a slope of -160dB per decade.

>>1934383
curious why do you need that kind of roll off? elimination is not usually the goal of filtering.

>>1934383
>From what I've read, it's extremely difficult to realize analog filters higher than 8th order, that is, with a slope of -160dB per decade.
its extremely difficult to do filtering even over the 4th order because you start to introduce more noise and parasitic impedances into the filter than you get attenuation ...once you get to -60dB attenuation thats usually less than the total noise floor for any use case and is pointless to do it

you use filtering to reduce the power of of incoming signals at certain frequencies or frequency bands so they become insignificant for the signal power and use case and are sized as such

if you want to remove signlas you use differential subtraction

>>1934383
Dude do you even realize what -160dB even means? Are you an audiophile by any chance?

>>1934382
>you draw the circuit,
>you use KVL laws or whatever you need to take out the equations and variables
>you set them in a readable format
>you make three matrices
>A matrix is the values(number_of_equations)x(number_of_equations) sized
>B matrix is the variable matrix with one column and rows as much as there are equations
>C matrix is the results matrix(behind the = sign) that is one column and rows are number of equations
>you use whatever preffered mathematical way of multiplying two matrices AxB=C
>you pull out the variable matrix and pick whichever value of variable yo uwant

pic related on the left are the results as A(values), C(results) , B(variable values after calculating ) matrices going from the top from the equations from the right

also you can do it by usoing symbolic variables if you dont have all the values but you basically do the same thing

>>1934411
he does not..like 90% of people that dabble in electronics dont intuitively understand what decibels mean or that 3dB drop is two times as quiet

Please help a brainlet understand a coil.
As far as I know the electric current always takes the shortest path because less wire is less resistance.
So why does the current go around through the coil instead of finding the shortest path through the copper and towards the exit?

>>1934417
Yeah I was thinking more on the automatic side of things ,I know how to solve linear systems.. Given a netlist, how do I solve it (starting with DC operating point and resistors and sources for example). I'm teaching myself Julia by making algorithms of stuff I find cool, now it is time for circuits..
Like how do you even describe a circuits as a matrix and how to apply KCL/VL to that? really got my noggins going.
Bottom plot has the wrong frequency values, I will fix it latter.
>>1934419
Because the wire is insulated..

>>1934419
>why does the current go around through the coil instead of finding the shortest path through the copper
The copper wire used in a coil is enameled, it has a very thin layer of insulation over it.
The current does take the shortest path, but since the turns are insulated form each other, the shortest path is around through the coil.

>>1934383
>-160dB per decade
>-500dB/decade
What the hell?

>>1934419
insulation
Winding coils are usually at least enameled (very thin coating) which is great to prevent shorting but with high voltages it is possible they will begin arcing. Though arcing rarely happens when used properly

>>1934411
>>1934383
I mean just in the case of you not being aware. -160dB is 1/(10000000000000000). You can't measure or much less hear anything in that range. Systems have parameters on what is acceptable to have a signal and what is other stuff. If you get within that then you are golden.

>>1934426
>>1934427
>>1934423
Ah, I see. Thanks for the help.

>>1934423
https://en.wikipedia.org/wiki/Nodal_analysis

try googling nodal analysis..and how net lists and nodes work in spice programs, thats basically how they do it

>>1934439
fuck yea h i remembered the name

topological analysis is the one that uses netlists
https://en.wikipedia.org/wiki/Topology_(electrical_circuits)
its specifically used in computer programs because its easy to translate into matrices and the circuits are combined into nets and nodes


idk how to do it step by step since we never really did that but it was mentioned that it was an old school method and its usually studied if you are making analysis software

>>1934439
>>1934445
Thanks!

>>1934308
>What sort of filter does 60dB attenuation across like a tenth of a frequency decade? Surely not an analog one?
The front end filter of your ADC *has* to be analog. What else would it be? Nothing. It's analog. Google "anti-aliasing filter application notes" and pick one. How perfect does your application have to be?

Hello. Does anyone here know what this circuit might be used for? Although I understand it is a common emitter with both active bias and active load, I can't see what its use may be.

>>1934411
>>1934418
>>1934427
Not him, but he's talking about the slope of the filter between 40kHz and 44.1kHz. What happens in the stopband could be a continuation of that slope, or a levelling out, depending on the topology of the filter.
The difference between those two frequencies is:
>log(44.1/40) = 0.042 decades
Which is really low. To get 60dB attenuation at the beginning of the stop-band, and ~0dB attenuation at the end of the passband, you'd need a slope of 60/0.042 = 1400dB/decade. While this is infeasible with more conventional topologies like Sallen-Key filters, I believe it's potentially possible with more exotic topologies like elliptic filters or whatever.

Now I'm not sure whether 60dB at the beginning of the stopband is really necessary, and it's certainly fine to have the top few kHz of the audio band being attenuated since you can just boost it back up again once it's digital without too much noise. From what I'm reading, the analog front-end filter is usually coupled with a digital filter after the ADC.

>>1934461
>anti-aliasing filter application notes
I'm looking at one now, but it doesn't have any specifics about getting such extreme cuttoffs.

>>1934471
Presumably, a mismatch in the two current mirrors will affect the gain, but it would also be dependant on RL, I think.

>>1934382
>>1934423
I took a class on this. It's simple to implement and I've done it in MATLAB. The books on this subject are a lot of complicated than what you need.
Learn about modified nodal analysis to turn circuits into matrices. Learn LU decomposition to solve matrices. I don't know Julia so I can't tell you how to create and solve matrices using that.
If you want a textbook, the standard reference is Vlach's Computer Methods for Circuit Analysis and Design. I'll link it here:
https://libgen.lc/ads.php?md5=1ef3d77090b180dada6358a7df8f962d

>>1934492
You also need what are called stamps so you formulate matrices correctly

>>1934471
>>1934490
What would happen if I tried this "open circuited", that is, considering RL infinite. Trying with a few numbers I get that the gain would be higher, but always less than 1. But I can't see the main advantage or use this would have over a normal CE

>>1934492
>>1934493
yeah i was talking about that but my textbooks are not in english so it wouldnt be much help ...we didnt do taht in class tho because it was considered not nesscesary to do by hand since its more algorythm based method so its suitable for software implementation

>Learn LU decomposition to solve matrices
i know python has numpy libraries for doing LU and probably also matlab has it ...it should be included in all math related libs since its one f the main things for solving equations with matrices so he doesnt even have to do it by hand

>>1934471
Could it be that if only one of the mirrors remain (that is, if it is only actively bias or only actively loaded) then either the collector resistance or the emitter resistance need to be very big?

>>1934493
>>1934492
based and circuit pilled

>>1934471
maybe a more stable CE without a decouplin capacitor?

My father-in-law has a rotary phone that I'd like to try and make usable on newer phone systems. Info on the bottom indicates it is a western electric 500mm. I don't have landline in my building, so I hooked it up to a "cell2jack" which is made to allow you to hook a landline up as a Bluetooth headset to a phone. I tried it with this phone as well as a brand-new princess phone that's your basic tone-dial. As expected, you can't dial with a rotary on this adapter since it's using pulse dialing, and any modern system is going to only work with tone dial. That's not my present interest. Rather, my interest is in the speaker volume. I've noticed that while the rotary has a nice loud ring, that it sounds very quiet in speech, and I can't tell if that's due to the adapter or something with the phone. The tone dial sounds much louder and clearer, and people can understand me much better on their end, so I suspect it is something with the rotary itself. My struggle, I suppose, is that I don't have any experience with rotary phones otherwise, and no one I know has a landline anymore, so I can't readily go and test the phone anywhere to confirm it works as I expected. So my questions are:
1. Advice on diagnosing a quiet rotary phone
2. We're they just always quiet/soft volume devices anyway?
3. Is there anything here in particular that would suggest "oh that's a problem with X" ?

Redpill me on DIY BCI. The way I understand it is basically like reading a remote control signal which is not that hard since you can buy the sensors etc. But the tricky part is how interpret the signal and map it to a specific command. So basically is it all about various software algorithms and filters and what not that process the signal from your brain and find certain patterns and execute commands based on those patterns? Simply put, what would be the purpose of it as a DIY project?

>>1934601
Did you get lost on your way to reddit?

>>1934592
The microphones used in old telephones were carbon mics, which want a relatively high phantom voltage and also draw a fair bit of current. If your circuit doesn’t have a higher-voltage PSU, or if it can’t supply enough current, then your microphone volume will be pretty low.
The speakers may also be a different impedance than more modern ones.

I’m interested to see how you’ll be converting pulse-dialling to tone-dialling.

>>1934601
The hardest part is probably the analog frontend and ADCs, it has to be noise-immune and probably requires some filtering too. But maybe you can just buy those as existing blocks. Once your MCU is reading that and feeding the data through a serial port, it’s a relatively simple software task to map what’s happening there to keybinds or whatever. There should be countless papers telling you how to parse and interpret the information.

Getting the electrodes themselves alongside all the other shit is probably gonna cost you.

I'm wondering about an easy way of converting a PWM output to DC to control a fan with no PWM wire, I saw a circuit that turns the PWM signal into a voltage using an op-amp but I figured pic related would work... the PWM pulses would allow pulses of current to charge the capacitor, and you could control the charging rate by adjusting the resistor... is this dumb?

>>1934297
The main problem with using coax for each signal is that the cable is usually quite stiff. This isn't a problem for the SCART end because SCART plugs are fairly chunky and the screen provides a strong mechanical connection. But the SNES end might not want to stay in.

>>1934649
Yes, a resistor + a capacitor will smooth out that PWM. What you’ve created is called a passive low-pass filter. But that resistor will dissipate a good fraction of the heat, so instead I’d use an inductor + capacitor low-pass filter.

But for an inductive load like a fan, PWM should work just fine without any filtration.

>>1932915
this is 2R right

>>1933457
>1.5 bits
how

>>1934260
I think he's taking the piss

>>1934686
Taking the piss? On 4chins? Color me shocked!

>>1934679
>>1.5 bits
>how

a bit is just a time period, so a tiny bit of extra circuitry can be used to count 150% as long. in fact, the hardware for counting half bits is already incorporated inside a UART coz they sample the input line at the half-way point of a bit-time.

>>1934659
Why? if the PWM signal is connected to a transistor for amplification & the PWM drops to the low part of its duty cycle the transistor should switch right off & you should get a tremendous back surge from the fan, & this would happen thousands of times a second. I imagined such a setup with a fan would self destruct, suitable for LEDs only.

>>1932125
More electrical than electronics but I have a motor used to lift a threaded rod I already calculated the torque needed to lift it up and down but my problem is the output torque of the induction motor is too high does any one know how I can reduce the torque also how to control the speed with a fixed torque ? form what I have seen for an induction motor a torque comes with a specific speed.

>>1934756
limit the current

>>1934756
VFD or series resistor. That is why induction motor is for the big boys and BLDCs are for us

>>1934716
Motor PWMs are run with a diode in antiparallel with the motor, which cycles the current back through the motor and prevents a voltage spike. You'd want a diode even if you were using an LC filter, just in case.

>>1934756
what kind of motor is it ..each type of motor has its own driving circuits ...if you have a big enough(heat dissipation wise) resistor you can just plonk that and limit the current but it might get quite toasty

Shill me a benchtop DMM, 500€ tops, no chink shit.

>>1934756
PWM of some kind. Raw PWM is probably good enough, but you could do constant-current with sufficient filtering and feedback. Direct constant torque is also possible by measuring the force on the motor's mounting points, but that's a right pain.

>>1934890
you dont need a 500€ DMM

>>1934902
>PWM a induction motor

>>1934890
You don't pick a benchtop DMM with price as your only metric, you choose what precision and accuracy ratings are acceptable, along with what ranges and units you need.
If you're considering a benchtop model, then presumably you already have a general-purpose handheld DMM; the only reason to get a benchtop model would be for the greater specs.

>>1934908
That's what a 3-phase/BLDC controller does, though smarter than a simple DC speed controller. For an induction motor, it's more efficient than using a VFD, and gives a reasonable approximation of constant-torque when open-loop. Though it's not necessarily as advisable from a practical standpoint.
I'm unsure if it's easy to provide analog feedback to a common VFD, and if open-loop you wouldn't be able to get much like a constant-torque, compared to PWM.
Ideally, you'd be able to vary both frequency and duty-cycle.

Phase-fired might work too, but I wouldn't try it.

>>1934908
>>1934826
>>1934756
>arent all electric motors induction motors
what do american mean by induction motor (AC, DC ,synchronous, asynchronous, perm mag,, brushed/brushless etc )

>>1934911
thats my point with >>1934905
if you need something you dont ask others what to buy, you ask others how to get what you need

>>1934912
Now this is low effort bait
https://en.wikipedia.org/wiki/Induction_motor

>>1934923
im not from an english speaking country and i always say the full name of the type of motor because i dont want to be caught up in semantics

>>1934926
>does not want to be caught in semantics
>gets caught in semantics regardless out of his own volition
Good job.

>>1934928
thats why i asked for an explanation so the next time i can be sure about what you are talking about

>>1934930
Oh sorry I thought you were trolling. Yeah, induction motor = asynchronous machines with squirrel cage rotors or regular rotors with the coils shorted

>>1934936
thanks

Hello /diy/. Basically what happened is that during normal usage of my computer, the pc suddenly shut down and from the GPU started coming out sparkles and smoke (the GPU is 3 years old and it's an msi gaming x 1080). What I discovered happened is that a ceramic SMD capacitor exploded, either for short circuit or other unknown reasons (unlikely). This happened probably because the gpu was used a lot and the pc stays on nearly 24/7. I believe that the capacitor is used along with a FET for voltage regulation for video memory, the section of the circuit it's in is in the 2 phase memory VRM section. I'd really be thankful if anyone knew about ways to check whether the internal layers are completely broken/shorted and whether it is possible to just replace the two capacitors, or additionally the fet too in case it was damaged by what also caused the capacitor to explode. I'm sending close up pics of the location along with a side view of the bulge that appeared following this (the charred protective black layer was removed).

>>1934966
Side view of the bulge.

>>1934967
ngl that looks fucking wasted

>>1934966
>>1934967
throw that out ..there is no salvaging that
if you see passives burnt out of the blue its almost always some active component that took it with them , one of your mosfets got an overvoltage dump into it and went out of regulation and burned your caps with overvoltage that shit is fucked

fyi capacitors and otherwise electronics in general preffer being always on if it is held at a (relatively) low and stable temperature, resseting and otherwise shutting down pc-s is nescessary due to bad and buggy software (you will see that servers and otherwise industrial PC-s are rarely if ever turned off if using a proper OS designed to be failure proof )

>>1934966
>>1934967
passives do not just yeet themselves out of existence like that, if you can not find out what happened then I do not think the repairs will last long. You could try it as the alternative is throwing out, but do not keep your hopes up

>>1934979
there is no fixin that anyways ...he got burned up traces and the worst thing delaminated board due to heat ..that is a big ass bubble and i higly doubt only power traces are in that sandwich of 10 layers

>>1934980
>>1934979
>>1934971
So not even MSI themselves can do anything about it (especially since the warranty ended anyway)?
What are my possibilities then? Any way I can earn anything out of this (like for the cooling system or the completely still working Core)?

>>1934983
you can sell it for parts probably ..some refurbishers can use some parts

but msi would not fix that..if you sent it in for warranty they would just give you a new one

>>1934983
I do not understand a lot about GPUs but it seems like trash now. But I guess you could salvage the cooling system and other useful passives like inductors (if the board has them) for other projects.

why are pmos in to92 so fucking expensive

>>1934996
bad package for MOSFET trannies ...its too big for low power things and not big enough(also no cooling) for high power things so no one uses them

>>1934308
There's a factoid in the back of my mind that says 44,100 samples/sec was picked so that you could fit the transition band (if that's the correct term) of an analog filter in those 2kHz.

There are some tricks that can be used:
The filter doesn't have to do all its work in 2 kHz; the transition band can be from 20-24 kHz, so twice as wide. This means that you'd allow frequencies from 22-24 kHz to "reflect" and create false signals, images in the 20-22 kHz range. This is normally a bad thing, but if it happens outside the range of human hearing, no harm done.

Psychoacoustic tricks. While young people can hear sound up to 20 kHz, no one can hear them well. We're just not very sensitive in that range, and what's there is mostly white noise anyway (sibilants, breaths, cymbals). I don't know if this is actually done, but I'd imagine you could have a transition band from maybe 18-26 kHz without anyone noticing. You'd get images of higher frequency content mirrored inside the nominal range of human hearing, but it'd all be hissing noises and no one could tell the difference anyway.

In practice what you'd use nowadays is a digital filter, like you say. Sample at 2x, 88,200 and use a much simpler analog filter with a transition band from 20 kHz to 60 kHz. Once it's digital, you use an aggressive digital filter and decimate down to 44,100.

>>1935068
or you just put in a normal low pass filter that starts at 18khz and once you hit 22 khz you have -30dB attenuation and anything above that is undetectable out of noise

>>1934967
>notices bulge
0w0

>>1934967
>bulge
The board delaminated due to heat, possibly due to a defect in the board. could be severe damage in the inner layers.
It's not worth attempting to fix, the board is toast.

>>1934979
Actually, cheap ceramic caps that are running close to their breakdown voltage do have a tendency to go kaput. Such a short time before failure may be due to thermal deformation of the ceramic itself.

>>1935068
>Sample at 2x, 88,200 and use a much simpler analog filter with a transition band from 20 kHz to 60 kHz
Pretty sure you have to filter from 20k to 44k, as 44k is your nyquist frequency. But yes that would be a rather easy way to do things, and the DSP side would be relatively simple.

>>1935078
>a normal low pass filter that starts at 18khz and once you hit 22 khz you have -30dB attenuation
That's 0.087 decades
that's -344 dB/decade, which is like 17 poles on a butterworth
don't use a butterworth

>>1935281
i was talking out my ass to be honest >>1935078
it going with the theme of the original poster

>>1935281
>>1935288
althogh you can get a 90db/decade slope with a cascade of three op amps

Why do appliances rated for 120v burn out if you plug them on a 220v socket? Why can't they just draw less current instead?

>>1935297
How would they know to draw less current? How would they sense the issue, and how would they act on that information?
You'd have to build something detect the voltage, and something to limit the current, which is extra parts, more cost, for a situation which would rarely occur. You'd have to have a transformer which could manage both voltages, which would be far more expensive, and pointless if you don't plan on using both.

>>1935297
because the components inside and the wire isolation are not rated for higher voltages

capacitors blow , transistors blow , diodes blow transformers heat up and burn due to the cores not being rated for higher flux values

>>1935300
W = A x V

>>1935301
So if all those components were made for higher voltage rating then the appliance would work on 220v? Why aren't all appliances made for a higher voltage then? So they would never burn out?

>>1935309
W = A x V
Stating this doesn't indicate how the circuit senses nor is capable of changing it's behavior without extra parts.

>>1935309
>So if all those components were made for higher voltage rating then the appliance would work on 220v?
yes thats literally the only difference

>>1935309
>Why aren't all appliances made for a higher voltage then? So they would never burn out?
because high voltage appliances require less current ...lower voltage appliances are lover voltage but they must use more current
components that can handle high current AND high voltage are more expensive than components that require one or the other

TL:DR because money

>>1935296
You can get better than that, 120dB/decade with three 2-pole butterworth sallen-key filters, more for that russian sounding one.
For a 20k-44k gap between the passband and the stop-band, you could get an appreciable -41dB with 6 poles of filtration from three op-amps like that.
Not sure with the intuition about how -41dB affects the aliasing however, other than the fact that it's 100 times less voltage.

>>1935297
Simple appliances (heaters, motors, anything with a mains transformer, any capacitive dropper) typically have a constant resistance, or at least something resembling one. P = V^2/R, so if you double the voltage without changing the resistance, you get 4 times the power being dissipated.
P = V*I doesn't mean much, as current is a dependant variable, not an independant variable like voltage and resistance.

More complicated appliances will use feedback of some kind in order to only take the power they need (usually with a switching power supply), but this requires some silicon 99% of the time.

>>1935322
thanks for the info ...although as i said i was just pitching in a bit because i never got into fitler design...but i definitely would not filter untill i did all the conversion so the aliasing would not be affected then filtering what goes out the ther end

>>1935318
I see.

>>1935322
>dependant variable
I didn't knew about this, you mean whenever i try to use those formulas for something i should always use them in terms of V and R?

>>1935330
>I didn't knew about this, you mean whenever i try to use those formulas for something i should always use them in terms of V and R?
no you use impedances , and think about the thing in terms of sources and sinks with input and output impedances with current and voltage requirements

go look up circuit analysis and n-pole circuits

>>1935330
It depends on the circuit. IF you aren't changing the appliance (R), and you change V, then I is fully determined.
Think about what happens when you max out your supply's current. Decrease R, I doesn't change, so V changes.
The dependency is about what is free to change, and what is fixed based on the whole (supply + appliance) circuit. No matter what you do, V=IR. If R is fixed, changing V must change I.
It so happens that it's physically easier to change voltage than current, but it's quire possible to change the maximum available I, exceed it, and witness the voltage drop to compensate.

>>1935330
Resistance (or impedance) is almost always a material property (exceptions exist like light-dependant resistors or FETs, and then there's non-ohmic materials), so is the first thing that you should assume to remain constant. Assuming such a model applies to your situation, that is. Almost all power supplies are constant-voltage when running in normal operation, with some exceptions like LED power supplies and some battery charge circuits. Benchtop power supplies often have a constant-voltage and constant-current mode, which is helpful. In a simple V=const, R=const. situation, it's only natural that current is independant. And in 99.9% of all cases you'll run into, power will be dependant too.

So it's context that tells you what's dependant and what's independant, and learning to read that context is an important skill in analysing any circuit.

Let's say I wanted a dc motor not larger than ~25mm x 15mm x 10mm, where would I go and look for that?

In General, how do I go and look for components, do you just use the search option of the various internet sites?

>The notebook died.
>Dissambles it.
>Found three of this blowed up.
How i can read it?
Exists any replacement?

>>1935441
bottom pic?

>>1935385
>In General, how do I go and look for components,

read first post in this thread.
however, motors are more of an Ebay/Amazon/AliExpress thing unless you wanna pay commercial rates.

hey guys i had an incandescent indicator/blinker/turnsignal on my bike and i replaced it with an led one but since the led puts way less load on the circut the relay doesnt go and the blinker stays solid or doesnt work depending on how i wire it, the globe is 12v/10w. if i wire a 10w 1.2ohm resistor inline with the led will this solve my issue? thank you

>>1935461
Post entire circuit, and what each component needs to turn on and turn off.

>>1935452
>bottom pic?
It's from the up.
For me it's a capacitor of tantalum.

>>1935464
https://www.manualslib.com/manual/977866/Aprilia-Rs4-125.html?page=92&term=Wiring+Diagram&selected=2#manual
here you go
really im just looking to simulate the load of a 10w 12v incandescent globe and need to know if a 1.2 ohm 10w resistor will do that for me

>>1935467
Yes, which is why I'm requesting a pic from the bottom

>>1935474
>Yes, which is why I'm requesting a pic from the bottom

>>1935474
Finally i found a replace slightly more bigger (250uf) and now works.
Anyway thanks for your attention.

ok so my cellphone's charge port broke
and the battery is fucked too
and i need some data off it for 2-factor auth to put on my new phone that weren't backed up
so i bought the teardown tools on ali and they arrived yesterday

first i wanted to try removing the battery and replacing it with a 3.5-4v cv power supply instead, spoofing the third pin which i assumed was a thermal resistor
but after some reading, i learned that because its apple, the third pin comes from a smart ti chip on the protection board of the lipo that communicates with the phone, not something i can ever spoof

so i went to trying to remove the charge port and solder some jumper wires to a 5v cv power supply, but removing those required removing the wifi antenna and a bunch of other shit, and the plug was really damn complicated

so instead i think i'll solder three jumper wires from the existing battery to the phone, hoping the connectors can stand the heat, and feeding a current-limited voltage into the battery itself
im not confident in my ability to solder onto the traces on the flat flex pcb/wire that the routes battery power to the connector on the main board
hopefully that keeps it alive long enough to boot it up while its still open, take the data off it (or back it up) and not have any of my wires break off due to metal fatigue or poor soldering

thoughts?

>>1935532
My confidence in your plan isn't high.
If the phone is communicating with the battery, it might not try to boot unless the battery reports it has capacity.
Just a guess.
You might do better by trying to charge the battery directly using a power supply. Go through the battery's terminals, not directly to the cell, so that the battery controller can count the charge. If it lets the cell charge at all.
if that doesn't work, see if you can get a replacement battery or borrow one from a donor phone.

>>1935461
>led one but since the led puts way less load

this is a common problem with some blinker circuits: the blink rate is determined by the load current. you can get a replacement LED-friendly blinker, or increase the load by putting a resistor IN PARALLEL with the LED. 1.2 ohms is kinda correct, you can try higher values so you dont get as much heat. trial and error.

>>1935461
>led one but since the led puts way less load

this is a common problem with some blinker circuits: the blink rate is determined by the load current. you can get a replacement LED-friendly blinker, or increase the load by putting a resistor IN PARALLEL with the LED. 1.2 ohms is kinda correct, you can try higher values so you dont get as much heat. trial and error.

and you def dont use a 10W resistor to dissipate 10W coz that's in the burn-you-fingers region. you'd go maybe 20-25W.

>>1935536
awesome thank you

Ohmbros, what got you into this hobby/profession? Genuinely curious and you don't have to answer if you feel like it's gonna clog up the thread. I'm trying to get other perspectives so that I might discover other interesting things to do into the future.

>>1935533
the battery currently has 3.7v in it, so it boots, but that voltage probably drops so quickly that it never boots completely because it black-screens on me before anything can start up
i can't charge up the battery first and plug it back in because the only place i can solder onto is the shitty connector
also i can't take the battery out because im retarded and broke off the adhesive pull tab, so i can only go to the battery's connector, not to the lipo itself

i don't actually have a power supply, so i was going to use an lm317 set to as close to the battery's voltage as possible, with some resistance in series with it, picrel

i also have a tp4056 lying about, would that be a better option?
the existing lipo already has a protection circuit on it so im guessing it doesnt really matter

>>1935536
surely theres a way around this that doesn't drain the battery senselessly, no?

>>1935550
as a hobby, you just pick a project or two you want to try and learn what you need to complete that project
as a profession, lmao good luck

>>1935560
>surely theres a way around this that doesn't drain the battery senselessly, no?

it's a blinker, so it's only turned on occasionally (or NEVER in my case - i use hand signals)

>i can't charge up the battery first and plug it back in because the only place i can solder onto is the shitty connector

why solder? surely there's a way to use needles, alligator clips, rubber bands or weights to just touch the right places. even just using your fingers to charge it for 5 mins.

>>1935550
>what got you into this hobby/profession?

some fucker gifted me a soldering gun at age 14. have been burning my fingers since.

>>1932125
How big of a rare earth magnet do you actually need to destroy a cellphone?

>>1935581
given they are solid state, very, very big.

>>1935578
the connector is really tiny and stuck in the case, so while not impossible to use needles, it would be cramped and would likely need to be for a lot longer than 5 minutes given the battery's condition
alligator clips are completely out of the question due to the size, and elastic bands and weights won't work due to the close quarters and such
i have no intention of using the phone again once i've got the data off, so fucking the connector isn't a fear of mine

I want to build a big fuckoff array of LEDs, put them in a picture frame and hang them on the wall. How should I heatsink this? Thermal pads on the back of the PCB to a metal plate? Vias in the PCB and the same? Just vias? Pads?

>>1935603
use aluminum substrate for the PCB, heatsink the back.

Ok if i have an analog video signal coming from an FPV drone, can I just generate an analog video signal of some sort of information (say, speed, altitude, etc.) as white text on a black background, and just add the two signals together before they go into my goggles?

NTSC or PAL, I can probably pick either. IIRC some anons here had experience with analog video signals.

the power source for this led light died and i need to figure out what is the voltage to drive that led

looking at the power source which takes 230AC as input power i don't see anything that would turn that AC into low voltage DC, but the output wires are thin red and black wires indicating the led is powered by something like 12DC.
So what could it be?

this is the led driver on that pcb
http://file1.dzsc.com/product/13/05/23/903845_162346374.pdf

>>1935640
an led is typically driven as current not voltage, the ic you linked will vary the voltage in order to meet the set current.
wire size indicates current not voltage
wire color indicates nothing do not rely on it
the ic is what turns the ac to low voltage dc, look in the doc you yourself posted to see the schematic.
page 3 gives the output current formula, you will need to find the value of R(ext) by inspecting the dead driver to determine the driving current.

>>1935640
test each part with a multimeter and replace the faulty ones.
start with replacing the electrolytic capacitors if you don't find it.
cut an old usb cable and connect the red and black wires. that's 5V.

>>1933486
make it into a transformer.

>>1935656
just a guess, but is that round component on the right a radial inductor, not a capacitor? The colour of the (ventless) top reminds me of a normal ferrite core, and the outside looks like it could be heat-shrink, maybe. While it's possible to use a TRIAC alongside a full bridge to dim a DC load, I doubt that's what they're doing, making me guess that it's a buck converter with a MOSFET instead. Can't read/find that part number. Then again, I don't see any control circuit, maybe it's open-loop or self-resonant? Or just under the goop? Or maybe the D2PAK itself is an integrated-FET switching controller IC? So many questions.

The existence of a capacitor that isn't a suppression capacitor, and a power resistor, makes me think it's more likely to be a capacitive dropper, even though the other parts tell otherwise. I'd quite like to see a reverse engineering of it's circuit diagram.

And no, that 510k resistor is not a diode.

>>1935627
>generate an analog video signal of some sort of information

no, that's some way complicated shit, creating and multiplexing synced video.
simpler would be to encode serial data as audio signals (like DTMF) and send them on the audio channel, if there is one.

>>1935661
>And no, that 510k resistor is not a diode.
lol
If it was mine I would have put more effort into it.
only thing I looked up was the rectifier.
I'm new to THT idk smd yet.

>>1935581
depends how hard you hit your phone with it ...but id say 0.5kg would be enough for an adult sized male

Anyone have ancient data on a motorola 1906-0218 bridge rectifier? Dated 8139. Looking for forward current and voltage. Nte cross online is no go. Moto pdf I found goes to '78...

>>1935740
what was it connected to ? bridge rectifiers are basically interchangeable as long as they are in spec by power(max voltage and max current)

>>1935740
is that a rectifier out of an old HP meter or something ? ...back then HP was government contractor and a lot of things were either classified or were made by order and didnt have common off the shelf equivalent and you will most likely never find a replacement unless you go salvage it out of another HP device that has a known good part

or get an of the shelf rectifier thats rated to whatever current that needs (heatsink and big leads would mean at least 10amps)

or measure it on a curve tracer (or by hand) and make one yourself (or order it to be made to spec)

>>1935788
It is a left over from a vast stash of parts I used to have. I very rarely fix much these days so the stash went away some years ago. I was a hoarder but I'm felling much better these days. I am attempting to replace a bridge in a 48vdc golf cart charger. Amp meter on charger goes to 15a. So... 25amp 100vdc or better is my goal. He dropped it off and will be back thru around 5pm. He lives 90 min away. I might be able to fix it today or wait a week for die hard overkill to arrive in a week.. His bridge has 3 terms along 1 axis with the positive 90°. I could have him order it and talk him thru doing it himself.

>>1935812
to be honest i would post that on EEVblog or on ebay someone might need that part exactly and you also might get a chunk of change while doing it (it will most likely cover the cost of a new one)

the rule of thumb is if you are not absolutely sure then get a new one (especially if you risk losing a quite rare part)

>>1935828
I understand that. I'll have him order a replacement. About 5 bucks for an overkill direct replacement on ebay from one state over. Rare part, I once searched for months for a rectifier tube for a zenith trans oceanic radio. I could not bring myself to modify something modern in its place.

>>1935281
>Pretty sure you have to filter from 20k to 44k, as 44k is your nyquist frequency.
You don't in this case. Frequencies above Nyquist will fold back and create images, which is bad. But frequencies between 40-60 kHz will fold back to the 20-40 range, and those images will be cut by the digital filter.
Anything above 60 kHz would fold back into the "audible band" so that has to be completely rejected by the analog filter.
(using rounded off approximate numbers to make it simple, pls no bully)

>>1932152
High current DC choke. I feel like i'm the solution to many problems but assholes insist on using capacitors.

I have a dvd player that isn't reading the dvds. It makes some noise as if the disk is rubbing on the tray every few seconds but it does spin and it also does recognise when there's disc inside so i think the laser is working at least to some extent. How can i test it and hopefully find the problem? All i have is a cheap multimeter.

>>1935932
you get a donor cd tray and swap it out ...aint no fixing old fiddly plastic bits that worn out (might be the motor bearings too)

Stripping a piece of wire is circumcision. Metzitzah B'Peh is tinning.
(you have to tin the jewish dicks, to make them stick)

>>1935947
this is not /b/ or /pol/ keep the thread clean of unnescesary racism...at least be humorous while you do it

>>1935560
>the battery currently has 3.7v in it, so it boots, but that voltage probably drops so quickly that it never boots completely because it black-screens on me before anything can start up

sounds a bit odd. Can you measure the voltage while it boots to see if it actually drops out? You might have something else wrong.
Anyway the 317+resistor might do it. If you've got nothing to lose you might as well try it.

>Agilent 54622 is LOUD
>install quiet fan with 15CFM less airflow
>scope runs silent but 10 degrees hotter
can't do it bros, I think this qualifies as abuse.

>>1936002
install 2nd fan and make your own vent.

>>1936015
>install 2 fans each half as loud as original
>???
>profit

>>1936016
2 x silent = ???

>>1934471
Could this work as a variable current source? Using R1 to control how much of the current from the up mirror gets sinked into the lower one, and how much goes into RL.

I am at my wit's end. My aquarium light uses T5 fluorescents, which afaik seem to work just fine. They start up and shine bright as all fuck, but then something craps out and they die out.
The complete situation is that 1 out of 2 outputs of the ballast doesnt seem to connect or even work and somewhere in the ballast you hear a chirping every time it turns a fluorescent on.
My problem now is that everything I find leads me back to "just replace ur starter" "replace the fluorescent" "it keeps flickering/it doesnt turn on at all," but it just flickers twice and turns off, besides, I dont even have starters.
What can it possibly be/what are some options I can look at or test to know more of what's happening?

>>1936113
Also pic related, the loud chirping thing. I have no idea what this is used for within a ballast, but I'm assuming it shouldn't be so loud.

>>1936113
>>1936121
Update update, I tried connecting the lamp via another way to the ballast as to exclude a faulty connector and that atleast seemed to fix the instant turning off thing. However, now it stays on for a few seconds, sounds like a bee nest and turns off again.

>>1936113
Sounds like you might be overloading your ballast(s), causing them to detect overheating and shut down after a few seconds/minutes.Especially if they’ve got a PTC thermistor at their mains input.

You said that one ballast wasn’t working? Then all the tubes might be running off the same ballast, causing the issue.

Either way I’d get another ballast just for testing purposes. Trying to troubleshoot the ballast(s) until they both work the same is another option. The whining coming from that coil is not something I could diagnose without seeing a circuit diagram, but if the circuit uses high-frequency starting it would make sense at least in the short term.

>>1932125
any one know the tipping point for voltage vs current for driving transformers. so like at what point does 1kv do better than 3.7 volts ? assuming 3.7 volts is 4 amps at what point is it making a stronger magnetic field at 1kv? same windings for primary assume 5,0,5 center tap. i know if you had enough windings on the 1kv even if it was 0.000amps it would be stronger but with frequencies thats not a possibility. it would never saturate the core of a transformer

i look every where and its just faggots pontificating about winding proportions and reminding every one that the secondary winding direction relative to primary winding matters . hell half the data you have to extrapolate from inductors . its not even for transformers but transformers are inductors

>>1936159
There really shouldn’t be any difference so long as your winding ratios are correct. Lower voltage means thinner insulation so I think it’s better for the most part than higher voltage; the copper thickness and number of turns cancel one another out. Easier to turn too.
If the winding ratios are correct, then the magnetic field in the transformer core should be the same regardless of voltage, hence saturation shouldn’t be an issue. Except for the extreme cases where you get hotspots of magnetism around particularly thick wire.

That’s not taking into account skin effect though, and I’m not sure how it varies by current and wire diameter.

Anyone in here done outdoor LED lighting? I'm thinking of replacing the strings of C7 garbage lights with some strips, but don't know where I can get reliable waterproof outdoor-worthy sets, and if it's still cheaper to get my own enclosures and shit for running them. I just need white.

>>1936240
There are some relatively common IP6x LED strips, and while I’ve no idea how good they actually are, they’re probably good enough for light rain, so long as the solder connections are properly isolated. Actual LED fittings definitely come in properly waterproof containers that double as heat-sinks, if you’re interested take a look at bigclive’s chinky street light reviews.

Hey there

Im trying to get some PCB's made through Aisler.
I have the board almost uploaded however I cant get the drill files properly.

Anyone have experience doing this, i only get .drl files from Utiliboard?

They are looking for two drill files:

>PTH Drills (plated holes drills) >project_name.drills_pth.xln PTH Drills (plated holes drills)

>NPTH Holes (non-plated holes drills) >project_name.holes_npth.xln

Not sure if anyone here knows the answer to this, but here goes:
Does anyone know to indirectly heated cathodes for vacuum tubes are made?
More specifically: I can't find much info on how the heater filament is insulated from the cathode.

>>1936271
Pretty sure they’re heated through IR radiation from the heater, no contact at all.

>>1936272
That's mostly my understanding as well, but it's pretty tight in there.
If you look closely at a tube you can definitely see something coating the heater.
I see some sources claiming that aluminum oxide is what's used, but I can't figure out how it's applied.

>>1936280
>something coating the heater
Likely just something dark to improve its emissivity, while itself not having much of a propensity towards boiling off electrons. Though the latter isn’t a necessary feature if the coil is properly placed on a floating winding. It’s also important to use a compound with similar thermal coefficient of expansion to the base metal, and a very high melting point, aluminium oxide may do this pretty well, at least I know it melts pretty hot.

As for the application, it’s possibly a vacuum vapour deposition process, like that device I saw on Thought Emporium. Or maybe it was Applied Science? It’s possible that it was applied by electroplating a metal then reacting it to create the ceramic, but this would be more likely to effect the filament itself.

I know they use a particular coating on the cathodes to make them boil off electrons particularly well (though it depends on the situation whether they use it or not), so it’s likely a similar process to how they get that stuff on.

Is there any way to find nice flexible IEC power cords?
I have a couple of old cords that are as soft and flexible as mic cable, they coil very nicely and don't kink.
Every cord I've bought new has a really stiff rubber insulation and doesn't coil as well.
Flexibility isn't really a spec that vendors offer.

>>1936280
>>1936289
Wait are you talking about that white frosty stuff? That's to prevent contact with the hole in the cathode through vibration, which is easier than propping it up with a piece of glass I guess. That coating isn't on the heater itself, just the copper(?) wires propping it up.

>>1936300
I've got some somewhat flexible vulcanised rubber cables (sandwich press), and some really flexible silicone cables too (DMM leads, soldering station cable), though those material markers alone don't imply any particular stiffness. You should be able to find cables with sufficient information in their spec sheet about durometer and such, but you'll likely have to terminate them yourself. I think PVC is never nice and flexible, and whatever "TPE" refers to can be a bit fragile.

>>1936289
>if the coil is properly placed on a floating winding.
That seems really difficult. There's not much space inside the cathode.
It has to be resistant to being bumped as well.

>As for the application, it’s possibly a vacuum vapour deposition process
I'm a bit worried that might be the answer.
Vacuum deposition is a bit beyond my reach.
>It’s possible that it was applied by electroplating
I've though about that.
Haven't had the chance to see if aluminum will electroplate tungsten wire.

>I know they use a particular coating on the cathodes to make them boil off electrons particularly well
I've kinda got a process going for applying calcium oxide to the cathode.
Calcium carbonate will dissolve in carbonated water, which the cathode can then be dipped in and let dry.
Heating the cathode then causes the calcium carbonate to decompose into calcium oxide.
Barium oxide and strontium oxide are also used, but I don't really feel like fucking with those to be honest.

>>1936303
>That coating isn't on the heater itself, just the copper(?) wires propping it up.
Are you sure about that?
It looks to me like it goes all the way up.

>>1936306
>That seems really difficult
I meant electrically floating. As in, not referenced to any other part of the circuit by virtue of being powered through an independant transformer winding.

Metallic aluminium might have some adverse effects at those temperatures if some is still left after oxidising/anodising it. And IIRC aluminium can't be easily electroplated at all, at least not in aqueous solutions.

A vacuum deposition chamber isn't actually that difficult to build, the glass dome (or mixing bowl) and vacuum pump are relatively common parts to buy (you'll need a vacuum pump anyhow), and you just need a ring magnet or two and some heating shit. Watch this and see if it's within your range:
https://youtu.be/Cyu7etM-0Ko
It's also possible that a non-aqueous redox reaction left the coating, like a thermite reaction.

>I've kinda got a process going for applying calcium oxide to the cathode.
Neat to hear!

>>1936308
Red is the frosty white stuff I'm referring to, yellow is what I'm assuming to be copper wire or some other wire with low resistance, and blue is what I'm assuming to be kanthal or tungsten or some other resistive metal that can handle sufficient heat. The white coating is definitely not on the copper wire or on the tungsten.

I'm not sure what exactly your other image is showing, though it looks like some thin wires with PTFE or maybe ceramic insulation.

>>1936336
>I meant electrically floating
I though that was the whole point of using an indirectly heated cathode.
If I wasn't going to isolate the heater I'd just go with the much easier to make directly heated cathode.

>https://youtu.be/Cyu7etM-0Ko [Embed]
Ok, that's pretty neat.
I might have to give that a go.

>pic
Oh, I think you've got it backwards. The thick wires (blue on your pic) are the feed-through leads and the wire with the white coating is the heater.
I'm pretty sure the yellow is part of the heater filament too.
The heater wires go into that circular hole, which is the base of the cathode.
In the other image, that's definitly not PTFE.
It might be ceramic, or it might be aluminum oxide, I'm not certain.

>>1935441
>have trash, what do?
dump it.
>>1935932
>have trash, what do?
dump it.
>>1936113
>have trash, what do?
dump it.

Next.

>>1936343
>I though that was the whole point of using an indirectly heated cathode.
I was assuming the purpose was for geometry reasons, to be able to have a cathode that's not a coil shape. Also potentially noise reasons, from the 50/60Hz on the heater, causing the voltage source for the cathode to be a bit more complicated.

>I think you've got it backwards
Shows how much I know about triodes.
>definitly not PTFE
>It might be ceramic, or it might be aluminum oxide, I'm not certain.
Aluminium oxide is a ceramic, and those solid smooth white (non-flexible) pieces almost certainly are made of alumina. If that (or other alumina parts of a vacuum tube) is what your google results were referring to, then the coating on the heater itself may be something else. I thought it looked a bit too frosty and imperfect to be alumina, but maybe it's just applied differently. Perhaps a sludge of alumina that slowly dries would be sufficient, similar to your slaked-lime coating method. I suspect the solid alumina insulator pieces are sintered together.

>>1936303
>but you'll likely have to terminate them yourself
that's the problem. I can get all kinds of cable stock, but the connectors are so damn bulky when they aren't molded directly onto the cable. Not a good fit for a power cord that lives with a piece of portable equipment.
TPE is "Thermo Plastic Elastomer", a very generic term.
I think the older cables I've had were made of natural vulcanized rubber. Maybe old stock is the only way I'll ever find premade cable assemblies made with such nice rubber.

>>1936159
>any one know the tipping point for voltage vs current for driving transformers. so like at what point does 1kv do better than 3.7 volts ?
what ?

did digikey ruin their parametric search? i'm looking for a boost converter and the parameters are all unsorted

What I want to do is have two potentiometers, one for 80% of the gain one for 20%. I need to limit the current going into IIn. The problem is that if FADJ goes negative, I need to limit the current into IIn further.

Basically in logic terms: IF(VFADJ > 0) -> normal resistance. IF(VFADJ <= 0) -> increase resistance as a function of VFADJ. I've been looking at different kinds of voltage controlled resistor circuits but I'm not sure what I'm doing.

What's the best way of limiting current into a pin if another voltage on the board goes negative?

>>1936659
>What I want to do is have two potentiometers, one for 80% of the gain one for 20%
Use an op-amp buffer in between the two gain stages, and select resistors to set how effective each pot is

>>1936668
Forgive the stupid question, but what's the advantage of having an op-amp?

Wouldn't any error from the potentiometer be multiplied by the op-amp? E.g. Pot has 1% error, gain of 2 on op-amp is now 2% error. Op-amp itself presumably has an error, e.g. 1% - so I'd have a 3% error.

>>1936679
An op-amp has arbitrarily high input impedance and somewhat low output impedance, meaning it will stop one stage from altering the next. The error of an op-amp is going to be much less than any potentiometer, whatever that means. And I’d suggest using them as buffers, not as gain stages.

So I'm still at the hobbyist stage where I'm comfortable laying out my circuits on perfboard rather than PCB. Although I've got a PCB etching kit tucked away somewhere, from what I see of PCB etching services you can only buy them in bulk (e.g. 10x PCBs). To those of you who use these services: what do you do with your spares? Do you only get basic module PCBs that are applicable to a wide variety of circuits?

is there a way to put capacitors on a 7 segment display to prevent flicker without using a fuckton of caps? also how do i physically connect all those caps for multidigit displays?
or just avoid that shit and get a faster driver(will still blink just faster)

>>1936754
What are you using to power them?

>>1936754
an arduino
if i switch faster it doesnt blink but something else happens, cant into words so heres a pic instead

>>1936765
Assuming you're also using a driver IC, place a decoupling capacitor across the power&ground pins of the chip. It might also be that your breadboard pins are touching underneath the plastic holes; Give them a gentle jostle and see if it improves / worsens the problem.

I'm currently using 3x2-digit 7-segment LEDs for a timer and I found that the arduino's 5V output couldn't reliably power the LEDs. If you have a separate power supply (e.g. benchtop PSU), power the IC and LEDs off of the separate power supply (so that the only connection between the display & the arduino is the signal outputs).

>>1936679
>Wouldn't any error from the potentiometer be multiplied by the op-amp?
resistor and any other passive compnent tolerance (or error as you state it) means that its value might not be exact as the manufacturer states it (10k resistor might be anywhere from 9500 to 10500 for a 5% tolerance margn) but if you measured it once and it said 9500 it wont change its value(maybe a little bit due to heat)

your potentiometer here is an absolute current limiter so there is no "error" to be had because you fiddle the knob until its set to a value you want...also using a buffer stage will allow you to set just how fine you want your fine and course adjustment to be

>>1936750
> To those of you who use these services: what do you do with your spares?
i personally work as a contractor that makes devices so most of the time if i order a PCB i need it in bulk anyways ...but if i order something as a prototype its either something that can be used unversally (like my own MCU breakout board or something that can be reused in multiple ways depending on waht i solder in like Hbridge power stage and can chose between half and ful bridge or comms board with multiple protocol chips and connectors

usually etching or ordering a PCB is saved for a final stage,most of the time its faster and easier to solder everything with wire bits in air or on a flat copper PCB (or protoboardPCB)

used to etch them by hand for end stage prototypes but its literally cheaper to order if you really need them and then use the spares as coasters or you give it to a friend or some kid (i keep in contact with my highschool teacher and donate him all my spare boards and designs i dont need so kids can learn soldering and electronics)

>>1936750
While I don’t use those services, it’s handy to have spares in case you mess up soldering. They’re usually done with SMD components to cut down on size and cost, so it’s both easier to mess up soldering and good practice in general.
Making the boards multi-purpose can help too.

But in general, getting PCBs fabbed is more of a “limited run production for testing” sort of thing than “1-off iterative prototype”, because of the shipping times and such. So it doesn’t really replace perfboard/manhattan/routing. I would definitely consider getting a bunch of SMD adapters for manhattan/perfboard/breadboard prototyping however.

Also I think some companies will let you have non-duplicate boards so long as you’re using up enough total FR4.

>>1936754
You’re multiplexing them, right? Increasing the frequency that you drive them should fix that as far as the human eye is concerned. A capacitor (and diode) for each segment would get annoying quickly, at which point I’d suggest you use 74HC595s instead. Or equivalent latching SIPO shift register.

Using capacitors will increase the effective turn-off time of each segment, which by the look of >>1936765, you’re already having problems with. Unless you have a way to actively discharge the capacitors (read: nightmarish current spikes), or you’re fine with the display changing numbers slower than you’re doing at the moment, it won’t look any better that your fast test.

So use a bunch of 74HC595 ICs.

>>1936776
>i keep in contact with my highschool teacher and donate him all my spare boards and designs i dont need so kids can learn soldering and electronics
That’s cool, man. I’d like to be in a position to do that in the near future.

>>1936754
vomitwire is usually called a bus
>>1936765
>cant into words
usually its reffered to as ghosting
personally i never saw that really happening , you should first check your code because an arduino should be fast enough to drive any 7 seg display...you might have delays or hangups on some parts of the code or you are refreshing the display too fast and your IC-s cant handle it or your ICs are getting false signals over the wires due to interference
also as anon said decouple your driver ICs and 7seg dsiplay with 10nF or 100nF caps

>>1936777
>because of the shipping times and such
JLCPCB usually makes and ships stuff for me in 3-4 days max ...in my small town thats literally faster than ahving to wait on my local ee shop to restock on photoresist boards and chemicals

>>1936779
to be honest i never even though about it before doing it ...the boards are cheap enough to just be given away even if you are a hobbyist if you really feel like thats something you want to do ....i was used to only in country board printing and that was hella expensive and every board was accounted for and reused even if it had issues or bugs (many a board had cut traces and soldered wires to re-route them) but once you went form 100€ for 5 boards and a week minimum wait time to 10€ for 5 boards and 3-5 days of wait time it quickly becomes expendable

>>1932152
I think we're all resistors here

electronoob here,
I bought all the parts for this impulsively,
Am I going to blow up or shock myself if I try to build this?
https://www.instructables.com/DIY-Bench-POWER-Supply-circuit/
Should I learn more before I attempt to assemble it?

>>1936817
So long as you don't short anything. If you're really worried wear thick non-conducting gloves (e.g. rubber kitchen gloves).

>>1936817
>>1936818
Also, The fuse should do its job before anything horrific happens.

>>1936818
>>1936819
Thanks goys,
I'm not really concerned with how the components/circuit works like I usually am, I just want to build a power supply to play with smaller components easily and learn about those.
I understand vaguely how it works because the guy explains it, but I'm not sure I should trust the chinkshit at 120v AC
Anyways if you don't hear back, you know that I fucked up

>>1936823
smaller circuits*

>>1936823
>chinkshit
that is just a buzzword for people that are angry at the fact that chinese do everything for cheap and no one can compete
>trust it at 120vAC
120V is just painful and at those low powers you would have to try hard to kill yourself..in fact you should (and probably will) get zapped at least once to see what its like

>t. i work with live 220V and got buzzed a ton of times

>>1936840
As much as I agree with you. it doesn't help that you're desensitised to it. Same way brickies work up high without their harnesses on.

>>1936842
You dont get desensitized to it dumbass it aint sucking on hot peppers...and i getbuzzed every few years not all the time...if you are not wet or barefooted and close the circuit over yourself through the heart its not that dangerous unless you hold on enough to get electrolysis to happen and get sick (which at those power levels if a solid ammount of time not a quick zap because you accidentally touched an exposed wire)
and its not a fear to get over it like people that work on heights

>>1936840
Putting it across your heart, especially with moist hands or sharp electrodes, could be quite dangerous even at 110V. And you never know if you’ve got a weak heart. So I’d still not recommend getting hit by any mains shock. Smaller shocks, say across the hand, can have a surprising amount of current going through them, and this results in burning around the contact area. About as advisable as putting your hand on the stove to see if it’s hot.

>>1936856
Yeah and as i said doing any of that especially with a pcb circuit requires you to put in quite the effort to kill yourself

>>1936860
>printed circuit board circuit
for me its the LED diode

>>1932125
Is there something I can buy to replace the batteries in a device with a wired power pack?

>>1937040
>wired power pack
What does this refer to?
It already has batteries in it, or do you want to replace a mains supply with a battery supply?
What is the thing, how much current or power does it need, and at what voltage?

>>1937040
>replace the batteries

a typical power pack delivers about 1/10-th to one-half of the power used by the device. that is, you charge it for ten hours to get one hour of run time. if you wanna replace the batteries, you just need a much bigger power pack that can deliver as much power as the device uses.

>>1936308
The cathodes of receiving tubes consist of a sleeve of nickel alloy coated with a compound of alkaline-earth oxides (Ba and Sr, usually). Inside is a tungsten heater wire insulated from the cathode with Be0 or alundum (aluminum oxide) ceramic insulation.

That's all I have. Cathodes were a well-kept trade secret. The patent applications might give some clues as to how these coatings were applied. Books tended to focus on how to use them instead of how to make them..

>>1937064
>alundum
never heard that used before

>>1937074
https://www.abrasivematerials.saint-gobain.com/products/micro-aluminum-oxide/electric-grade-38-alundum

>>1935640
So i found out the problem, one of the traces is ripped off.

So i connected it to AC and measured the output and multimeter shows 296V DC which is obviously bullshit, but also 11V AC which is more legit, so it's probably a generic 12V led since leds can run on recitifed AC if i'm not . mistaken.
Since there is no transformer, i wonder if the 11V output is save to touch.

>>1937113
the output is perfectly sage to touch, if you touch the 12v output wires, just do NOT touch something like a water pipe or metal computer case when you are touching it, since it looks like a capacitive dropper, so it's only 11v relative to itself.

>>1937113
>>1937121
actually scratch that, unless you know what you're doing you shouldn't touch it

>>1937083
Oh it's a brand name referring to a particular mix/size, not a general term like alumina. Neat to know. I'd apologise for filling up the thread with off-topic garbage, but hey we're already past bump limit.

>>1937113
>which is obviously bullshit
Not at all, rectified 240V mains can be as high as 340VDC or so, after filtration. Because sinusoidal 340Vpeak gives the same power output as 240VDC (with ohmic load), hence why we call it 240VAC. When we say 240VRMS, the RMS stands for root-mean-square, which is shorthand for the method of integration by which the voltage is calculated. For a sinusoid, this just means a factor of √2 = 1.414.

The ripple at the end will be seen as AC, and while 11V is a bit large, at that voltage and for an LED light it's passable. Basically what you've got is a ~300VDC with 11VAC superimposed atop it. Note that the 11VAC might be inaccurate due to the waveform not being sinusoidal but rather a bunch of capacitive ramps, which some DMMs may not work with very well.

In other words, yes you definitely have rectified mains, and no it isn't isolated so watch yourself.

Post nice, well-lit, square pics of both sides of the PCB so I can reverse engineer it, because it's an interesting cirucit. That or find something identical on bigclive's channel.

>>1936872
Yeah i know but what can you do its the way you talk these days ...i was writing a paper and my professor insisted i write LED diode instead of just LED because "LED is what coloquially we call a modern light" and i was like wtf you talking about man

>>1936855
I meant that you're desensitised to the fact that it happens rather than actually getting shocked. Someone who doesn't get shocked regularly is probably not going to be so blase about getting shocked.

>fear
HSE (UK OSHA equivalent) mandates the wearing of harnesses in high areas (e.g. working on rooftops). There's always a story every few years about some retard plummeting to his death.

>>1937182
>meant that you're desensitised to the fact that it happens rather than actually getting shocked
That has nothing to do with the fact that its not nearly as dangerous as people make it seem ..i got desentised to the fear that its a boogeyman that acts like a finger of god ....get some good insulated boots and dont stick your fingers in very high current applications that have huge ammounts of energy and you wont be dead(line voltage rails on consumer devices are very low current and dont thave enough energy to be effectively dangerous unless you are trying to get yourself killed)

>HSE (UK OSHA equivalent) mandates the wearing of harnesses in high area
Again that has nothing to do with what i was saying...desensitization can happen in fear indiced anxeities like a fear of heights and can be mitigated by repetition and practice ...of course you should wear a harness and any safety available even if you were born on mount everest

>>1935512
can you really trust it?

>>1937139
https://patents.google.com/patent/US1975870

This RCA patent describes a construction process for the sort of of heater-cathode assembly in your image.

Anyone got a good log scale of dB vs perceived loudness? I'd like to make a VU meter and for frequency bands but I want each LED in the stack to represent some non linear loudness level if that makes sense

>>1937366

if you google it, every single one has this scale from -20 to +3, so just copy that in digital in however many steps you need.

>>1937064
>>1937083
>>1937364
Thank you, that's helpful info.
I hope I can get something working sometime later this year.

>>1937210
>insulated boots
In what situation does current ever flow through your feet? For the average hobbyist, we're going to be doing shit inside, standing on a wood or carpeted or concrete floor. If you're outside or on a metal grate, sure, but so long as you're wearing closed-in footwear, anything is probably fine to resist a measly 240VAC. Heck, even crocs are probably fine so long as they're not wet.

As for sticking your fingers in appliances, I'd be much more wary of high-voltage devices than high-current ones. Stick my hand directly across a 100mF capacitor, won't feel a thing if it's only at 24VDC.
On the other hand, I think many mains shocks happen when you have a relatively high impedance between you and the mains (like that time I shorted the open on-switch for a <10W appliance with my finger), so in that case I agree with you.

>>1937379
Not him, but that's kinda a weird scale, and I thought dB were already logarithmic. But even assuming it's a useful scale (it probably is), I wonder how they get the driving circuitry to respect that scale? Diode log amplifier? Actually because dB is already logarithmic, it would need to be two log amplifiers. Which would be a right pain to tune.
Unless the second logarithm is provided by variable springiness of the meter or otherwise by its magnetic geometry.

>>1937442
Crocs are never okay.

I have a pic related. For an SL-7 turntable. Someone is trying to jew people for one on eBay and I'm not buying. I also can't find something alternative because the pulley is proprietary.

It's a 4v dc pwm motor with a small rubber belt that controls a worm drive. It's clearly not working.

I took it out and ran 9v through it to see if it would get it loosened up and going. No go either direction and just clicks when the voltage connects. No whirring or sound coming from it otherwise. It turns by hand but feels stiff.

I don't see any way to pull the fucker apart to access the guts.

What do?

>>1937471
The brushes might be fuckered. If you can't take it apart, just swap it for some similar DC motor, doesn't have to be a match, so long as it has the pulley on the end of its shaft.

>>1937474
No clue where to find something with a pulley like this to spec. I've google searches for days. The pulley doesn't appear the be removable from the shaft.

>>1937442
Rectified mains powered stuff will often pack an outsize punch for the voltage involved and will also be a shock hazard if you make contact with anything grounded unlike transformer powered gear (unless you ground the chassis). But yeah, it's only 170/340 volts DC.

I'm planning on building an extremely simple analog synth in order to learn about amps and other related parts, but I'm not sure about all of the details really. My plan is to start with a VCO module (controlled through an external sequencer) and a output module (headphones jack and a volume knob).

My sequencer uses 10V for control voltage, so I thought I'd use that throughout the entire system. I've found some VCO examples of 10Vpp at 0-10V, so that shouldn't be a problem. As far as I understand, the oscillator output will range from 0-10V as well.

However, I'm not sure how I'll drive the headphones. Can I just offset the input to -5V to +5V and then run the signal through an audio amplifier?

why tf is ltspice not giving me accurate values

what was that other free simulation software called? starts with an 'm' I think?

>>1937495
Do you want a really barebones VCO for testing other shit? Use a 555 timer on astable mode. It's pretty much "my first VCO" before you start fiddling with CEMs and even takes control voltages.

Although there are proper audio amplifier circuits, using a single opamp as a signal buffer will increase the input signal up to the supply voltage. A square wave will only require +V whereas saw, sin and tri will require +V & -V.

>>1937502
Fritzing?

>>1937518
>Use a 555 timer on astable mode.
Great idea actually. I already have one laying here somewhere.
>A square wave will only require +V
Sure, but you can't drive headphones just alternating between GND and +Vcc, right? I've done some minor DC projects, but this is the first time I'm touching AC. Or did you mean that I should pass GND-+Vcc to the opamp wich then outputs +Vcc - -Vcc?

>>1937526
I should have said that the 555 will only output a DC square wave that alternates between 0 and +Vmax (essentially a DC wave). To boost the output you only need "extra +V". An amplifier requires -vcc when its amplifying an AC wave (which is what most audio comes in). However, if you have the components then a 555 & a signal buffer won't take long to throw together as a starting point.

>drive headphones
Speakers generally only require the signal and a ground connection to complete the circuit. The volume of the speaker depends on the amplitude of the wave which is boosted by the amplifier.

>>1937495
for 0-5V CV, look at the VCO for the TB303. Very simple schematic, easy to adapt. You will need to learn about log converters and temperature compensation.
If you just want ambient drones, hex-schmitt triggers can give you 6 oscillators super easily

>>1937537
>square wave that alternates between 0 and +Vmax
Yup, that I actually understood. Thanks.
>Speakers generally only require the signal and a ground connection to complete the circuit.
Oh, I see. I thought you'd need -Vcc to "pull back" the voice coil quickly enough for the speaker to work. Just needing signal + ground simplifies things a lot. Cheers, mate!

>>1937541
Thanks.

>>1937556
No it always works, it just wont produce air waves of enough power to be audible

>>1937502
>fritzing
lmao

you are most likely not setting up your schematic properly if lt spice is not showing yo ucorrect values ..its basically a glorified claculator

otherwise the other software you are looking for is multisim from national instruments (you will have to pirate it)

Which one of these are 240v?

>1/2

>>1937643
>2/2

>>1937630
Is there a list or guide of what to look out for? I've tried adding every parasitic resistance/capacitance I could think of. I'm getting crazy discrepancies like 0.5V ppk outputs, while the sim says I should be getting 9Vppk