/diy/ - Do It Yourself

this thread's digits brought to you in part by the AD8538 low-power precision auto-zero opamp

I made an account with jlcpcb but cant log in to place an order. This happening for anyone else?

>>1698551
I can log in, but the pcb won't upload properly. It usually bugs out over the weekend, i guess when staff is low. I bet if you try tomorrow it will be ok.

what's the story in OP's pic? Also requesting to add w2aew to Op's list of tubers. He is a really good teacher and profilific dude.
https://www.youtube.com/watch?v=7nmmb0pqTU0

Looking at iron-core inductors for loudspeaker usage, something in the range of 2-3mH. I can't find much information on linearity and only DCR and nominal inductance value.

This link is about all I could find.
https://hificompass.com/en/projects/experiment/test-couple-iron-core-inductors

Is the no. 3 legit?

Question:
I am converting a van into a camper, and put in a solar setup. I had a 100ah AGM battery in with a 200W solar array, but the battery got cracked and started offgassing. I took the battery out without disconnecting the solar panels from the charge controllers. It sat like that for a few months.
Here's where the problem comes in: I put the battery back in to check everything, but my charge controller doesn't turn on and I don't believe the battery is charging at all. Now, I am HOPING that the solar panels and all the wiring etc. is fine, and it's just the charge controller that is fucked up. What normally happens when you have solar panels hooked up to a charge controller with no battery?

>>1698590

Yes. Those two large SMT capacitors on the end are for smoothing the output of the circuit. They can be small because it oscillates at anywhere from a few hundred kHz to like 1.5MHz. If you mean specifically the claims about electrolytics are tantalum caps, those are also technically valid. They're not usually a concern, though. A well-made electrolytic, specced appropriately, will last decades, and tantalums need to be abused fairly badly to go nuclear.

That being said, that looks suspiciously like a DC/DC buck converter module I bought a while ago, and it was utter shite. It had huge quiescent current (like 200mA if I remember correctly; advertised at <2mA) which caused it to overheat even with nothing actually connected. It didn't fare any better when driving a load. I'd look for you to be sure, but I threw all of them out like a week ago.

>>1698559
https://en.wikipedia.org/wiki/Boeing_787_Dreamliner_battery_problems

>>1698609
>tantalums need to be abused fairly badly to go nuclear
apparently they're more delicate than elcaps, almost as pissy as lithium batteries. a touch too high of voltage and kaboom. a bit too much ripple current and kaboom. the only real place for them anymore is under FPGAs when you really do need 1000µF on your 0.9V rail, directly under the balls
>shite
what voltage source? those big cercaps do cause massive inrush currents. given a long enough input wire and an input voltage close enough to the max spec, the overshoot can hit you hard
https://www.analog.com/media/en/technical-documentation/application-notes/an88f.pdf
also, automotive power is pretty hostile
also, you weren't back-feeding its load, were you?

>>1698567
>caoacitors aren't dangerous in a functional appliance
if we're still only talking what is allowed to back-feed out of the power inlet. inside a device, you proceed at your own risk. two words: dielectric absorption
>just a precaution
that depends entirely on the circuit and the amount of energy in play. high-current low-voltage devices like laptop power bricks often run themselves down to a safe voltage over a very short period of time and/or contain a bleeder resistor to help discharge the caps. low-current high-voltage power supplies such as in CRT cathode supplies can't afford to waste the power on a bleeder resistor, so you bring your own, on a long stick
but see https://en.wikipedia.org/wiki/Dielectric_absorption for a spoopy wrinkle

>>1698653
>>tantalums need to be abused fairly badly to go nuclear
the problem is when they go they go short instead of every other cap in existence ever that go open and that you probably wouldn't even notice missing.

>>1698653
>a touch too high of voltage and kaboom

Overvoltage is arguably the single worst thing you can do to a cap. I'd consider that "fairly bad" abuse. It's "maximum" voltage for a reason.

>what voltage source?
18V tool battery. It was for a 3D printed adapter that would turn Ryobi tool batteries into massive USB power packs. Don't get much smoother power than that. Ended up getting some nicer modules that were larger and cost like 3x as much, but they actually worked well.


As an aside, I found one that didn't get thrown out. It was one of pic related, not quite what you have pictured. No idea if the problem was poor design or fake/bootleg components, but, either way, 0/10 would not use again.

>>1698684
>why is there a fuse on my capacitor
don't you always put a fuse on a bomb?
by "a touch too high" I mean 95% for a long time. modern practice is to replace with ceramic if possible, or derate working voltage to 50%. feels bad man
>MP2307
looks about like the front-page typical application. dat efficiency vs. load current graph doe
my luck with MPS converters has been mediocre. among other things, MPS converters have the sad misfeature of a very limited abs max voltage on their enable inputs. even on the MP2456 converter, designed for Vin=50V, the max voltage allowed on the enable pin is 6-7V

>>1698513
Because those caps are charged through a diode (bridge) that prevents the current from flowing back through those pins. The same applies to a capacitive dropper, there's no current path from live to neutral or back again that isn't impeded by a diode or two.

Anyone know if you can buy SIP or DIP R2R ladders from alibay? They'd be pretty handy for working with MCUs (possibly through use of a 74HC595) but I can't seem to find any.

>>1698737
SMD ain't going nowhere, grandpa. you want useful, join the 21st century already

>>1698738
t. missed the point
I can’t find SMD ones either. The only resistor arrays I can find are either 8P4R SMD arrays (4 unconnected resistors of the same value) and the standard SIP parallel resistor arrays with a single common pin, again all the same resistances. Searching for “R2R” only shows me expensive audio equipment with DACs built-in, no ICs or other individual components like what I’m looking for.

Such a ladder would also be combinable with an MC1459 SAR to turn it into an ADC, which would be handy for those times when you want mess about with digital signals without a DSP or MCU.

>>1698763
I wouldn't expect to find them in single packages today when there are (and were) more convenient, less fussy ways to DAC. simply use 2 Rs like I did the last time I built an R-2R ladder for my Atari ST's parallel port. slightly lower chance of non-monotonicity that way. 0.1% resistors are surprisingly cheap in 0603 packages at LCSC. order a $2 board at the same time, get $15 off DHL express

DUDE MONEY TREES KA-CHING
>inb4 lol money trees
nope, it's legit P. aquatica

Anyone know/remember what that camera input connector is called? I want to bypass the VHF in but the camera is the only other input there is

>>1698809
so-239

>>1698812
Fuck that was fast. Thanks man. Gonna try to get a so-239 to bnc converter for my Ntsc generator

New Logitech mouse PCB. What's the cm scale for?

>>1698828
Making sure that the chink pcb sellers don't undercut them on the important antenna length?
Silkscreening happens at the end of the PCB making process so it's not like it could be used a reference for anything.

>>1698812
>>1698814
Imagine a crowd-sourcing platform where questions are automatically reformatted and directed to the correct 4channel general for the fastest response. The platform would fail because half the time the answer would be "faggot".

>>1698609
>(like 200mA if I remember correctly
maybe you shorted something, or had a broken one, i have used many tiny bucks from chinks and never had a problem

I have a relay which can handle DC 24V 6A max, but i need to use it for 12V 10A, is it okay to simply use two relays in parallel?

>>1698737
>R2R ladders

Mouser, Digi-Key, Arrow, Newark all sell 'em, but i suppose you prefer dodgy chink parts.

>>1698851
>is it okay to simply use two relays in parallel?

yes, but the faster one will die pretty quickly, and his slow brother will join him in relay heaven soon after.

>>1698737
>R2R ladders

Mouser, Digi-Key, Arrow, Newark all sell 'em, but i suppose you prefer dodgy chink parts.

>>1698851
>is it okay to simply use two relays in parallel?

yes, but the faster one will die pretty quickly, and his slow brother will join him in relay heaven soon after.

>>1698833
<>>1698737
>R2R ladders

Mouser, Digi-Key, Arrow, Newark all sell 'em, but i suppose you prefer dodgy chink parts.

>>1698851
>is it okay to simply use two relays in parallel?

yes, but the faster one will die pretty quickly, and his slow brother will join him in relay heaven soon after

>>1698833
>"faggot"

you misspelled ''nigger''

>>1698737
>R2R ladders

Mouser, Digi-Key, Arrow, Newark all sell 'em, but i suppose you prefer dodgy chink parts.

>>1698851
>is it okay to simply use two relays in parallel?

yes, but the faster one will die pretty quickly, and his slow brother will join him in relay heaven soon after

>>1698833
>"faggot"

you misspelled ''nigger''

>>1698859
but 'faggot' is not racist

>>1698865

you would insult 4chan but claiming it is not racist?

>>1698865

you insult 4chan by implying it isn't racist.

>>1698859
I'm just surprised there isn't a single one there. As far as DACs go it's one of the cheapest methods and very arduino friendly, both of which sound right up some ali shop's alley.
>but i suppose you prefer dodgy chink parts
I prefer free shipping to the middle of bumfuck nowhere without having to build up a $50 order. I barely manage to buy $10 of shit every few months from ali, though this has gone up a little recently.

Then again, I am the guy who still hasn't got his shit from 45 days ago thanks to Yanwen. Not sure if I want it delivered now or in 16 days' time so I can get refunds on it all.

>>1698896
With individual precision resistors and generic resistor arrays being so cheap a single package R2R array is just a labour saving thing which don't mean shit when you've got children and robots assembling your boards, so why would china care.

For some part count saving and likely some advantage in matching over using all discrete resistors you can use n 4 packs of R + 2R to get 4n-1 bits.

>>1698851
Probably not. Whichever one closes first will get the full 10A at the point the switch closes. Whichever one opens last will get the full 10A at the point the switch opens. The points where the switch opens and closes are what causes wear. A single 6A relay might be able to safely sustain 10A provided that the current was below 6A during switching. But if you're actually trying to switch 10A, two relays won't be any better than one, because one of the two relays will always take the full brunt.

And that assumes that the 10A load is an "ideal" 10A load. Capacitative loads have a current spike when powered on, heating elements draw substantially more current when cold, inductive loads create a voltage spike when powered off.

tl;dr: 2x 6A relays won't work for switching 10A. A single 10A relay might not work for switching 10A if the load is in any way non-ideal.

Will this work as a voltage level translator for 5 MHz SPI communication?
I tried using a level shifting IC (CD4505) but it only works up to about 500 kHz

>>1698997
why are you spending time asking instead of building it and trying it out, faggot

>>1698833
VTR guy should have started his own thread particular to the project, like that FuG guy did, instead of spraying it across ten /ohm/ breads. a bit late for that now

>>1699007
I really didnt realize this project would take this long. The biggest problem is that when I order parts based on /ohm/ advice it takes 4-5 days to arrive so I have nothing new to post until then, so by the time I have more questions it's already a new thread.

>>1699009
nah you're cool VXbro. just saying it would have been nice to have the project itself packaged up in one nice little bread to show off to the grandkids or whatever

>>1699027
ah its all good. I got a thread going on videokarma.org about it, once I get this thing fixed for good, I'll make a comprehensive bread about all the things I did to fix it, to help out all those other VX bros out there

Hello, most audio effects I see arround the web use LDRs(easy to do) or expensive fets or some other autistic thing. I want to try using pulse frequency moduation (PWM is simpler but the circuit is more complicated). My question is, how do I demodulate the signal?

>>1699082
look up frequency demodulation

>>1699082
>implies other designers who have decades of experience have overlooked a simple, obvious and cheap solution
>has literally no fucking idea what they're doing
/diy/ the post

>>1698814
The male connector for an SO-239 is a PL-259.

Are there any UK component suppliers that still have physical catalogues?
Would be comfy to have one sat on my desk to flip through rather than having to stare at a screen.

>>1699091
No no no I'm not implying that. Sorry. FETs are perfectly fine, that is why they are used. It's just that they are expensive and I have the shit to make a PWM/PFM circuit but no fets
>>1699088
eh. I think I'll go pwm and just use a lpf. thanks anyway.

>>1699130
>have the shit to make a PWM/PFM circuit but no fets
if all you need to do is multiply one signal by an LFO, fine, but in something like a phaser or VCF you need to vary an RC time constant (but see also switched capacitor filters)
>PFM
mixing products and feed-through weren't bad enough to filter out at fixed-ish frequencies with PWM. I wouldn't ask for more trouble
>I have the wrong junk, wat do
join the US Democratic Party

>>1699128
RS will post a print catalogue by request, Rapid have an abridged print catalogue for the education market.

>A clearance other than standard Customs entry is required for this shipment.
is that my punishment for ordering chink brand MCUs

I'm thinking about buying a general purpose benchtop PSU, namely an HP 6267B (0-40V 0-10A)

My question is: how would you test the psu for being able to put out the 10 amps at any voltage? i dont have various 10 amp devices handy to hook up and make sure theyre getting enough current.

>>1699128

you can order them at luddite.com/electronics/catalogs

>>1699178

it has current limiting so a short circuit is a perfectly acceptable load. or use the 10A range on a DMM (as a load) to double-check the panel meter.

I didn’t want to make another thread and this one was pretty relevant- I have to ground a tow module in the back of my car. There are a couple bare screws in the frame already sticking out on the interior but I think they’re just for the taillight housing. Should I just drill a hole in the expose metal frame, sand it up some and screw down the terminal connector with a metal screw? I don’t really want to run a ground wire from the front strut tower but I guess if I have to?

>>1699180
>Yes yes, be a good goy and do all your shopping online it's much cheaper for- I mean er, ahem, it's much better

>>1699188
How is shopping by mail order with a catalog substantially different than shopping online? What the fuck is even your point?

>>1698601
The charge controller usually works off the battery. If your battery is connected and the charge controller doesn't turn on, check the battery for voltage. Odds are it's fucked. If you are 100% sure the battery works, and you didn't ground your charge controller, then your charge controller might be borked. They do usually say to disconnect solar before battery.

Currently have 215Ah lead acid deep cycle with 300W of panel, but I ran a vent hose from the battery box out the floor of the van to the opposite side from the exhaust. If AGM's gas like lead acid, why go with AGM?

>>1699182
>it has current limiting so a short circuit is a perfectly acceptable load. or use the 10A range on a DMM (as a load) to double-check the panel meter.
man, it literally couldn't be easier. ty.

I made this EL lamp driver. it's pretty much just the HV859 application circuit with adjustable frequency. It's rated for drive frequency up to 1KHz but it will push to 4K without trouble.
not as bright as I hoped, maybe 210v p-p just isn't enough. or maybe it's just this wire.

>bUt I cAnT bReAdBoArD wIth SMD

>>1699300
>shipping a breakout PCB from china takes HOW long?

>>1699291
Don’t those guys work best with a resonant driver of some kind? Pretty sure a matching circuit of some kind would be able to just drive it with LV and get the high voltage out the end too.

>>1699302
>>1699300
>need to breadboard 20 different smd components just one time
>spend 10 hours soldering them to breakouts just for one time use and then desoldering them back for smd use
yeah nah
smds are just a fad that can never replace tht

>order replacement part to fix my vive controller vibration
>3 weeks later literally still not shipped
>check seller location, Hong Kong
I really wish there was some (re)seller for replacement parts not in China. Literally $3 replacement part I need to replace and solder new wires to make it work again but getting the part is the biggest pain in the ass.

>>1699353
>smds are just a fad that can never replace tht for prototyping
FTFY
For PCBs where you'll need enough components that it's worth buying them separately from the THTs you used to prototype it in the first place, or otherwise PCBs where there isn't a THT equivalent then I think it's fine to go SMT. Especially if you're making more than 1 of a board. But for one-off circuits with only a couple dozen passives and an IC or two from the bin of hundreds of THT ICs you already have, there's no point in SMT.

Can a stepper really have almost double torque at exactly the same current as another stepper that is half length? or am i getting chinked?
So basically at exactly the same electricity provided i will be getting double the torque basically for free, just because the motor is longer? i don't want to accidentally break the ohms law and get into trouble or something

>>1699369

motors dont work on ohm's law. they work with magnetism, so a bigger heavier motor has room for bigger magnets, and bigger electromagnets, and so greater power.

>>1699375
But with same current??

>>1699376

yeah, coz flux is proportional to current and turns. more room for turns, more flux.

>>1699376
You could theoretically have a motor that provides finite torque without any current flowing, torque doesn’t imply power. Torque at speed however does.

On a side note, electrostatic motors anyone? Having speed proportional to current and torque proportional to voltage would be pretty interesting. But for the level of voltage necessary to get it to move, I imagine your commutators would make plenty of noise. Unless you went for a BLDC with electrets instead of a brushed (universal), but the switching method would need some thought.

Can we get 3D printer high-permeability and high permittivity filaments yet? It would make prototyping motors much more accessible.

>>1699309
Resonant drivers are larger in size and it's harder to have arbitrary frequency control. I wanted to try out one of these single-chip solutions.
If it works well enough to roll with, I'll move to a smaller no-leads package and some 0403 components to make it really tiny.
Gonna try it with different wire and see how it does.

>>1699353
>10 hours
if it takes you more than 15 seconds per joint, git gud
>desoldering them back
SMT passives should be considered consumable. sorry about your lucky resistor

>>1699359
>not keeping SMT passives in stock
how many THT resistors does a reel of 3000 SMT resistors buy
>keeping THT ICs in stock
"But when I became a man, I put away childish things; now that I am become a man, I have done away with childish things."

>>1699387
>You could theoretically have a motor that provides finite torque without any current flowing
How? Torque without power, sure. But how can you expect to have any flux without current?
In ideal conditions you can get speed with no current, torque with no voltage, but torque with no current? Doesn't maje sense to me.

>>1699387
https://www.csc.fi/web/elmer
go ahead,create and patent your briliant idea

>>1699375
>>1699375
>motors dont work on ohm's law.
This is the dumbest thing I ever read on my life. Please kill yourself
>>1699369
more magnets, a smaller airgap, improved core material.. There are millions of things that can make that happen, or you are being chinked

>>1699466
With an electrostatic motor.

>>1699502
Oh thanks for the link!

>>1698538
Noob here working through make: electronics.
Can someone please explain to me what this nigga is doing with this circuit?

It's a circuit for exploring audio frequency filtering using tones generated by a 555. I don't understand why he's included a LM386 power amp. I assume the amp is only going to be able to amplify up to the supply voltage, and the 555 outputs at at supply voltage, so why is the amp needed?

>>1699600
>I assume the amp is only going to be able to amplify up to the supply voltage, and the 555 outputs at at supply voltage, so why is the amp needed?
current

>>1699616
This, the 555's output impedance is awful. Not to mention it actually only outputs up to 3V or so. It also has terrible quiescent current and bad pinout, if talking about the 555's general shittiness is on the table.

>>1699616
Oh wow yeah of course, sorry for the brainlet moment.

Can an amplifier be thought of as a "variable supply current resistor" in a weird way then? Since a speaker is basically an open circuit when you "amplify" a signal what you're really doing is just allowing current from the supply to reach the speaker in a pattern controlled by the signal? The original signal itself isn't actually "amplified" as such, it's just used to control the "release" of coulombs to the speaker, like a button controlling a floodgate?

>>1698538
Does anybody know good pan-tilt motors? I've done some preliminary research into some drone stuff and robotics stuff, but I'm not sure if what I've looked at would be easy to control. I'm trying to make a visual tracking system, so something that I could get the rotational position out of and support a camera would be necessary.

I do have experience with embedded programming and computer vision, so at the high level I think I'm capable, I just don't know about a good motor.

>>1699629
a better first approximation of the LM386 is a voltage-controlled voltage source that just happens to be built to pass relatively large currents and is programmed for a 20x voltage gain by default. usually, power concerns are treated separately from signal concerns
>a speaker is basically an short circuit
fixed that for you
>isn't actually amplified as such
but that's exactly what amplification is: controlled passage of power between a source and an output, according to an input

>>1699659
if it uses stepper motors and you drive them properly, you would, or should, always know where the camera is pointing

>>1699661
Thank anon, that all makes sense.
>fixed that for you
Shit yeah, that's what I meant - "open" in basically every other context means that stuff is allowed to pass through so I always mix that up in my head.

>>1699661
Thanks for the response. I'm looking more for a whole assembly where I just control the motors rather than making it myself. I'll try to find something with stepper motors specifically, but I'm having difficulty finding something other than security cameras.

>>1699659
I think you should buy an old PTZ surveillance camera, maybe you can get one cheap if the camera part is broken.
Just tear the pan/tilt stage and motors out of it, tear the camera out of that, put your own camera in, and use your own motor controllers to drive the motors. depending on how old it is, it would have permanent magnet motors or stepper motors, you can drive either one with an arduino shield or something like that.
Point is, the hard work here is already done for you and you can get it really cheap. then you can focus on your code.

>>1699503
>>motors dont work on ohm's law.
>This is the dumbest thing I ever read on my life. Please kill yourself

being inductors, a motor's impedance varies with frequency. ohm's law has nothing to say about frequency so it doesnt apply.

motors, being generators, create a back EMF which varies with speed. ohm's law has nothing on speed or back EMF so it doesnt apply.

Every child should have one of these kits it totally tells you how the world works.

>>1699691
>ohms law = specific version of field theory to fit lumped param model
>what is
>motor circuit models
>not being pedantic with what is field theory vs what is ohms law
>complex impedance

Just finished a guitar pedal, I made a tremolo. The fallout asthetics is a byproduct of being poor.

>>1699502
>cmake error: no fortran compiler found
what year is this

>>1699727
what field do you think you're in

At what point do you transition from making input impedances as high as reasonably possible and output impedances as low as reasonably possible to matching impedances? Is there some gradual frequency threshold? Or perhaps when you swap from operational amplifiers to single transistor circuits?

>>1699742
>At what point
about 1/16 of a wavelength

>>1699727
Eh, those'll buff out.

>>1698538
Where can I get a circuit design for a 2-5 mile FM transmitter? I have no electrical engineering experience.

>>1699783
>transmitter
That's probably illegal

>>1699787
Don't worry friend. I have the required licenses.

>>1699789
What frequency are you going for?

>>1699791
Any amateur band. I'm out in the middle of fucking nowhere so it doesn't really matter.

>>1699794

for an on/off signal, or for audio? 1-way or 2-way communication?

>>1699795
Frequency modulation does imply analog, since frequency-shift-keying is a more correct term for the corresponding digital mode. But who knows.

>>1699794
you ain't gonna be able to do it without using up a lifetime supply of luck and having someone's hand up your ass
just use a CB, good buddy

>>1698538
Sorry if this is not the right thread but I don't want to make another one just for this. I'm trying to pull apart an electric motor and I'm stuck at the point in pic. Anyone know how to get this part off?

>>1699824
that's what /qtddtot/ is for
>questions
>that
>don't
>deserve
>their
>own
>thread

>>1699826
alright ta

I want to use electrolysis to restore some old cast iron pans. I did it with a bike chain and it worked great.

I'm contemplating the idea of building a PWM driver with varying frequency/duty cycle. I'm trying to look up on the Internet how a pulsed circuit will affect the job.

I understand that plain old DC will give me a constant pull. I've been wondering if some pulsed signal will act like small thugs, making the job more efficient.

Any thoughts?

>>1699844
>I've been wondering if some pulsed signal will act like small thugs, making the job more efficient.

It'll just make it less efficient. Ideally, you want as much current as possible with as little voltage as possible. Excess voltage mostly just starts doing electrolytic separation of the water. If the total inductance of the circuit is low (relative to the modulation frequency) PWM would just be pulses of excessive voltage followed by gaps of nothing. If the total inductance is high, there's no difference between PWM and clean DC.

Does Jaycar have an electronic components bundle/kit full of transistors, relay's etc? Or something on ali express?

>>1699824
bearing-puller or something along those lines is my guess
a mallet might also do the job, depending on the other side

>>1699864
I think there might be a bag of transistors or two that could be considered worth getting, but if you have the time to wait go for ali instead. An assortment including BC5XX and 2N390X complementaries along with some other ones would be a good pick.
For capcitors, get a bag of ceramics and a bag of electrolytics, perhaps a few X and Y caps if you're into mains stuff. Relays, probably expensive enough that you want to buy them for each project instead of in bulk in advance, same with toggle switches and some connectors.
Breadboards (and solid-core breadboarding wire?) always buy from china.

also jaycar has an online catalog, just search through that and compare prices, there's not going to be any difference in quality.

if ssr has reaction time 10ms, is that fast enough to create dimming effect for leds with pwm?
i basically want to fake dim the leds by triggering the ssr with pwm and then change the duty cycle to make them seem more or less bright
but with 10ms reaction time, the pwm freqeuncy will have to be only like 30hz maybe?
would that be enough?

>>1699896

if we're talking LEDs for lighting, they have a phosphor coating to change blue to white, which will smooth out the peaks. still, if you move your head fast, you should see some strobe effects.

>>1699902
well i just realized it is already blinking at 50hz so 30hz is too much lower and yeah i can already see the strobe at the mains 50hz when i move my arm while looking at it, but this won't be a led for regular room lighting, but rather for effects

>>1699896
>>1699903
Just use a MOSFET or three? The LEDs themselves run on DC after all. Might require a bit of hacking though. Also if it is an AC model, you could check if it works with a TRIAC dimmer, which might be slightly Berger than 30 Hz.

>>1699906
>Just use a MOSFET or three?
i need isolation and the lights are on mains so that is too dangerous to not use ssr

i guess i will try it out and see how it looks

>>1699907
>i guess i will try it out and see how it looks
MOSFET won't work on an AC load at all

>>1699903
>already blinking at 50hz
There's a secret component called 'bridge rectifier' that doubles the flicker frequency from 50 to 100 Hz and allows you to use a FET for switching.

whats worth keeping from old discarded circuit boards? sometimes I remove large capacitors and heat sinks even though im not sure ill even have a use for them. is it worth the time to remove certain components or should I just throw all this shit away. its a bit of an eyesore but I cant help but feel there are parts I might be able to use

>>1699824
what you are planning to use that duster motor for?

>>1699935
>heatsinks
>inductors
>xformers
>exotic components
>leds are always good I guess
But mostly magnetics and ICs, like the power management/supply ICs or micros

>>1699945
>exotic components
like mangos and coconuts?

>>1699954
Idk m8 that ultra sonic recombobulator you've been looking for or something like that

>>1699965
there is a VERY fine line between scavenging and hoarding
watch yourself fren

>>1699864
relays don't usually come in assortments, no. large varity of contact forms and pinouts. plus they're more than 5 cents each
transistors, resistors, caps, inductors, those all come in assortments, just search

>>1699902
no, it won't. the blue still shines through the phosphor and contributes to the perceived color

>>1699907
google PC817 optocoupler

>>1699935
thank it for its service and see last two lines of OP

I think Ive slightly stepped over the line into hoarder territory. time to start throwing shit away.

come and get your $10.10 500MHz+ USB 2.0 *HS* microcontroller dev boards faggots
https://www.mouser.com/ProductDetail/NXP-Semiconductors/MIMXRT1010-EVK

Complete beginner here, no wiring experience.

I'm trying to make a chord, on one end is a 4 prong NEMA L14-30R and the other end has a 3 prong dryer.

What the fuck do I do with the extra ground wire?

>>1699966
I've learned that the hard way, I only get cool shit with motors, spiny blinkies and expensive things now. Otherwise I'd live in a ferrite castle.
>>1699972
the first step is to recognize the problem

>>1700020
/ohm/ RULE 0

fuck you USPS, i just want my crystals

I dont get how antonio from amazon can get in somehow just fine but ups and fedex drivers are so fucking useless in this regard. Ive had that happen and it pisses me off

>>1700041
pack of wild nogs roaming around when they tried to deliver?

>>1700065
its weird because the unit has a front mailbox and a camera security, they could easily put a tiny padded envelope of 3.57mhz crystals in the box

Any buck converter to turn 12V 3A into 5V 6A in a reduced space? (1 inch square)
SMD is preferred

>>1700060
because the greek merchant doesn't believe in safety standards

>>1700071
you're looking at 3.5mm wide traces on 1oz copper to carry the output current with a reasonable temperature rise. even 6A is a big ask if you don't want to use a separate switching transistor. the inductor will consume the majority of the 25x25mm you would like to fill. what do you even need 30W for anyway?
find yourself a parametric selector for buck converter ICs at TI, MPS, or other vendor of your choice and get psyched up and ready to lay out your own

Electrics ≠ electronics.
Does anyone know what this thing is?

>>1700084
RGB clock with 127 WS2812B LEDs, was hoping to use a standard 12V 3A PSU and use a buck inside, I'll see if I can find a 5V 5A PSU, since at worst only 80% of LEDs will be on.

>>1700091
consider using two or even three bucks

>>1700089
Sperre = barrier ~= line filter

>>1700108
Thank you.

fuck. order too much stuff from LCSC, they give you a normal top-opening flappy box instead of a front-opening flippy box
BUT WAIT

>>1699891
This is the other side. I'll try beating it later today.
>>1699943
Not sure they're good for much are they? Was just going to have a look.

>>1700179
Beating the shaft from the back-side (the side in your pic) could result in the entire rotor along with that front bracket coming out. Be careful though, ensure that nothing will be damaged as it falls out of the motor.

>>1700179
>>1700179
>Not sure they're good for much are they? Was just going to have a look.
They have a shitload of torque and need a load otherwise they'll speed up and die themselves.
Also
>if you look the field windings are in series with the rotor ones
>if you cut them, feed DC into the field ones and spin the motor you have a generator.

>>1700091
that's pretty neato actually. yeah use two bucks. don't need to parallel, as long as all the grounds are solidly tied together 5V above ground is still 5V above ground
you cool with SOIC-8 with exposed pad? consider the MP4462, it is a pretty serious converter, maybe even more than you need or want. inductors sufficient for a 3A load at 2MHz are available in a 6x6mm footprint. for freewheel dioes, SS34 is available in an SMA package (SMAF is sexier). I don't think we've hit 12x12mm yet but I haven't considered track width. still more than enough room for the feedback network and bootstrap cap

Maybe I should've asked here first >>1699431

Also the project I'm thinking of will use piezo speakers, which I heard lose impedance as frequency increases. Mine are 8ohm and plan to use them from 15khz to 28khz, they're generic ones and I don't know how low their impedance is at those frequencies.

>>1700253
Voice coil speakers are inductors. Voltage controls the volume and current*voltage (as in all things ) power.
Piezos = high pass filter, they have very high impedance pic related is the response of a piezo guitar pickup I used >>1699723. I used a FET input amp (tens of mega ohms input) and a 10M resistor to the reference. (and 10 was a bit low)

>>1700276
Using a lower resistor "tone sucked" and made the bass response shitty, the sound was metallic. And as a rule of thumb piezos are usually "high" voltage devices.

>>1700277
It says one meg but I simulated and built with 10, the picture was taken at a different time.

>>1700280
Pic is the model of the piezo I used.
And you have to understand that power = / = loudness
https://www.ocn.org.au/guide/245-loudness-and-sound-intensity-power

Anyone know how much force a dome switch can take before warping out of shape? I need a small flat-ish button that can withstand a few hundred pounds of force. Or should I just use a membrane switch?

>>1700341
Perhaps a force-sensitive resistor would be better? Those things are designed to undergo high forces and are pretty thin. But you would need some sort of simple sensing circuit.

Why does it need to be thin?

>>1696945
>>1697024
Oh you were right. Turns out the litte sprung hook tip I was using somehow brought the voltage down to a couple dozen mV's.

Pretty sure those tips only purpose is to keep the wire firmly grabbed and they're supposed to be "transparent" right? Or am I missing something? What I found weird was the frequency reading was accurate. Maybe i'm using it wrong though, I just put it on top and applied pressure, I see no screws or anything else.

>>1700276
Piezo microphones tend to be used far away from their resonance frequencies, piezo speakers tend to be used near their resonance frequencies.

>>1700341
clearly your mechanical design is shite. fix that first

>>1700344
yes. sometimes they don't establish a very good mechanical connection on the probe side. they also aren't meant to bear mechanical loads on the hook side and should be laid so that the weight of the probe and cable is borne elsewhere

Man, whenever I see electronics stuff I get flashbacks of college. Sitting there poring over a bunch of equations. What a tough time, just grinding through these things and spending late nights getting it done. Is there a more fitting designated place of "the torture chamber of torture chambers"? I think not. lol Differential equations make me short circuit, honestly.

So I saw this project
https://github.com/ploopyco/mouse
and thought building my own mouse may be a cool idea as first major project.
Now the proposed is using a typical atmega32 to convert the signals, running at 16Hz. What if used another MCU with a top speed of maybe 60Hz? Would it even make a difference, besides me just toying around?

>>1700347
a silver lining to not being degreed in the field is not having had to systematically pig bottom to every sort of heavy mathematics

>>1700349
nope, HID is still restricted to 1ms minimum report interval

>>1700013
>MIMXRT1010-EVK

> so NXP comes out with a board that's somewhat arduino compatible, and sell it for $10
> presumably trying to compete for the attention of the ''cool arduino kids''
> but instead of a giving it a human-friendly name they call it a fucking MIMXRT1010-EVK
> Suzy, for our project should we use an ''Uno'' or a ''Mim-scrhit1010-evak''?

somebody needs to fire all the humanity-hating robots in marketing!

>>1700367
cool arduino kids are too shit stupid to USB HS. fuck 'em, more for autists

>>1700349
>and thought building my own mouse may be a cool idea as first major project.
you should to useful projects that will make your life easier, not a mouse that will never be nearly as good as the dollar store mouse, that is how you get burt out on this hobby

>>1700367
the problem with cattering to arduino autists is that like 95% of arduino tier projects can be handled by $3 uno, or $0.5 attiny, or $1 esp01 if you want wifi and there is really no real incentive to buy anything more expensive

Do chinks sell liquid tin where you dip pcb into and the exposed copper gets tinned?
i can find any on ali

>>1700353
>nope, HID is still restricted to 1ms minimum report interval
Thanks, I didn't think of that.

>>1700389
What kind of attitude is that if you can't build stuff for fun now? Most of what's being useful is done analogue in my house anyway.

>>1700409
Liquid steel would also be nice, and solid mercury.

>pic related

Hey guys I this oscilloscope at my workplace, its collected dust for what looks like decade. I cleaned her up and moved it by my bench. Any use in keeping it around, or some interesting projects I could work on with it? Seems like a waste to have it sit there like that.

Fuck. There seems to be a hole in my design.
I put in fresh batteries yesterday and today they are dead.
To turn on the circuit, the button is pressed and the load (arduino) boots up and pulls the mosfet pin high and then user lets go of the button and the arduino does arduino stuff
this part works flawlessly
then when the arduino finishes it pulls the mosfet pin LOW and this opens the mosfet and kills all power to the circuit, until it is started back up with the button again
In theory it works, when i press the button the arduino works and then it turns it self off and the power leds on it turn off as well so it seems like it's off
But there must be some energy drain somewhere, since it killed my aaa batteries overnight even though it was supposed to be off

any ideas?

>>1700439
>Most of what's being useful is done analogue in my house anyway.
wew gramps

>>1700409
a bit hazardous to ship, unfortunately
if you can't get a prepared solution such as MG Chemicals offers, you can search chemistry YouTube for electroless tin plating or immersion tin plating and find directions to make your own, which may be easier for you to get hold of (or not)

>>1700460
did you think to calculate how much current your 10k pulldown is pissing away?
get a fucking ammeter you moron

>>1700486
>did you think to calculate how much current your 10k pulldown is pissing away?
that would be exactly 0, as in zero
when the arduino is off, no current flows into the output pin and thus into the gate, and even when it is on the current is so small the batteries would last weeks if not months
so thanks for nothing, moron

>>1700488
>when the arduino is off, no current flows into the output pin and thus into the gate, and even when it is on the current is so small the batteries would last weeks if not months
but, in FACT, it's on. the circuit is always right, which necessarily means your theory of operation is wrong
plot some voltages on the schematic. measure it. prove it
https://e2e.ti.com/blogs_/b/analogwire/archive/2014/01/17/back-powering-why-are-the-lights-on-when-the-power-is-off

>>1699824
>>1700179

In the first photo your posted, the brace that goes across the motor shaft appears to be held in with rivets. Those need to be drilled out and the rotor will come out with the brace.

>>1700277
You used a 10uF capacitor at the input, of course the bass is going to suck, that is a basic high pass filter.

Heu guys i am wanting to use some punch downs for a project but, i dont need big ones you commonly see, i searched on ebay, and digikey but i cant seem to find them, single ones i can solder to a prototype board (or put them in a breadboard would be nice but closest im finding is pic related. Does anyone know what I am talking about.

>>1700573
Okay i think IDC connector is what im looking for, funny how it starts to show up after making a post

Im playing around with analog to digital converters and to get the maximum resolution Id need around a either a 1.8 or 2.4 ish voltage reference. I have no experience with references, where to buy, what to look for. Shunt transistors keep popping up.

>>1700638
TL431 is the standard jellybean voltage reference, add one resistor and you have a pretty precise source of 2.5V

>>1700448
It has an XY mode right? try this: https://www.youtube.com/results?search_query=oscilloscope+music

>>1700486
>ammeter
>in series with a resistor
lmao just measure the voltage across it

>>1700041
>two days later
>they havent tried to redeliver it yet
FUCK THE POST OFFICE

>>1700501
Piezos are usually a 100p-10n capacitor.

>>1700704
By this I mean, it doesn't matter. What matters is the input impedance. This is from 1n to 100u.
>>1700345
Thanks, did not know that.

I need UV lights to dry adhesives. I'm a newfag to diy stuff generally, can a kind anon help me out?

The adhesive needs:
405 nm @ 2000 mW/cm2 <1 second cure.
405 nm @ 250 mW/cm2 1 - 2 seconds cure.
320-450 nm @ 2000 mW/cm2 <1 second cure.


Where would I find suitable lights?

>>1700041
>>1700693
is that one guy still here? This is the crystal I bought btw MM49N (20pf). Its no HC6 but i dont know if that casing is still around anymore. It was cheap and figured it wouldnt take 2 weeks to get here. Should i order a different once and file for a missing items refund?

http://www.quartz1.com/price/techdata/MM49Q%26MM49N%26MM49O%26MM49S.pdf

>>1700749
yes, X(C)=1/(2*pi*f*C)

>>1700806
yep HC6 is a dead letter. people get cute sometimes by cutting the HC6 lid off of the base, removing the existing quartz element, soldering a stock HC49 encased crystal like your pic in place, and reinstalling the HC6 lid
but you can install it directly to the board. do not put any strain on the lead gaskets while forming the leads to fit

>>1700759
a UV nail "dryer" is great and ready to use

>>1700820
I know the complex impedance of a capacitor, what I'm trying to say is that a 10u farad or a 10 farad cap won't make much diference because they are in series with the piezo pickup, which can be seen as a small capacitor (10n-10p) in series with a large resistance. If you look at the bode plot a 100u cap puts the cutoff at 10Hz and a 1n at 20-30hz (which you can barely hear, if at all)

Does anyone know how those wifi extender sockets work? Most importantly, what kind of circuitry they use to receive and transmit data across the wires at mains potential. I have no intention of ever using them, but I figure it could be used to make a local IOT network that doesn't require your wifi to be running, just injecting a few kHz of low bitrate data, like for motion sensors or other monitoring systems. Which would also give me an opportunity to mess about with data modulation techniques and SNR juggling.

The receiving circuit is dead-simple, just a high-pass filter and some protection diodes, but I'm stuck on the transmitting side, especially since I'll be fighting against the output impedance of my local grid. Do I try to modulate current? Voltage? Do I use a second high-pass filter?

>>1700843
Pic related
>impedance of my local grid
which is mostly inductive, right? X(L)=2*pi*f*L

>>1700843
Oh and I also need to fight the substantial input impedances of all the appliances on the local grid, meaning the lower frequency I go, the closer I get to a frequency not filtered out by these appliances. So I should be looking to a higher frequency, even though it will mean getting closer to regulated radio bands. Of course, heaters and other purely resistive loads will always be effectively a dead short across my output, regardless of frequency.

>>1700857
Interesting circuit, but that receiving circuit with no filter and just a zener? Would you be able to read anything off that at all? Or is it just for mains frequency timing?
What frequency is that totem-pole driven at?

>>1700861

long time ago built a circuit from a magazine (pre-internet) that used 455Khz IF filter transformers to send and receive an AM signal thru the power lines. the trick, of course, is to select appropriate caps in series with the power line that pass 455Khz but block 60Hz power from entering your circuit.

>>1700861

long time ago built a circuit from a magazine (pre-internet) that used 455Khz IF filter transformers to send and receive an AM signal thru the power lines. the trick, of course, is to select appropriate caps in series with the power line that pass 455Khz but block 60Hz power from entering your circuit.

>>1700861
zero crossings are a bit quieter than the rest of the wave. previous pic was a transmitter only. Pic related is an example receiver
see http://zigbeedomotica.nl/x10-receiver/ for text
>frequency
the X10 standard calls for 120kHz carrier

>>1700868
this anon gets it

>>1700843
>a local IOT network that doesn't require your wifi to be running
this is a pretty great idea. any lights/fans you wanted to control would already be plugged into outlets. there's potential here for modding one of those cheap wifi outlets, you could put any micro you want in one of those.

p.s you'd be surprised how reliable those powerline ethernet devices can be. the tech works.

>>1700653
Wow that's awesome thanks!

>>1700868
Ah, I think my problem is that by going for 1kHz I was needing at least three layers of passive filters, each of progressively higher impedance. But going for 500kHz I'd have a whole 4 decades to fit in attenuation. So I could just use one ∏ or T LC filter, giving me -40dB/dec and get a full -160dB of attenuation, or 340Vpk down to 3.4µV. Which is really fucking good. Even just a -20dB/dec ∏ or T RC filter would drop it to 34mV, which may be sufficient, depending on the strength of the transmitter.
Not sure how that filter transformer works, but I'm assuming it's some sort of LC filter.
Also not sure how the ∏/T filters work when you're not matching impedances, just that if I want to transmit and receive through one filter I think I'd need to do it that way. I'll have to check that my inductors are good to 240VAC.

>>1700873
>zero crossings are a bit quieter than the rest of the wave
>previous pic was a transmitter only
Wait, you mean it was deliberately only transmitting near zero-crossing? That means you could ignore clipping destroying some of your data and recover some signal-to-noise. Kinda useless if you already have a high-rejection high-frequency filter, but definitely something to consider if I do want to go for lower frequencies.
So that receiver uses a low-pass LC with what should be plenty of rejection, but then you clip the shit out of it with D5 and D6 before amplifying? I don't think I understand it, but that link is certainly a great start, thanks.

Also what the shit, I always thought you needed a full-bridge rectifier (and hence an inconsistent 0V rail) to use a capacitive dropper, that circuit just blew my mind.

>>1700887
>this is a pretty great idea
Never heard that before.
>you'd be surprised how reliable those powerline ethernet devices can be
Yeah, I found the MAX2981 on google images (http://pdf.datasheetcatalog.com/datasheets/maxim/MAX2981.pdf)) and it or something similar might be usable as a general-purpose IC for this.

>>1700901

googles has multiple simple designs all ready to learn from or copy.

>>1700901
>only near zero-crossing
yep. imagine what a phase-angle dimmer would do to the desired signal if the signal were sampled at a point where significant current is available to be drawn. now imagine 50 dimmers up and down the block and a few brushed motors for extra lulz
>clipping
what, from the mains frequency? carrier frequency is 1000x the zero crossing frequency, 2000x the mains frequency. that's what, 33dB down?
>low pass LC
resonant tank, fren
>clip the shit out of it with D5 and D6
you know those shitty chink LC meters with the 16x2 character LCD? the voltage on the caps can get into the high tens of volts while powered from a 5V comparator
actual signal will probably be millivolt-tier on the mains, but strengthened by the tank with each carrier cycle. by the end of the 1ms burst it will probably be well strong enough to cause a noticeable change in collector current in Q2, on both the upswing and the downswing
>cap dropper
for a few mA it's not really a big deal to just burn the energy from the other half of the cycle. Asian Energy Star dad may disapprove
>matching
doesn't matter quite as much for just ringing a tank. X10 is shit slow (1 bit per zero crossing). with some attention to the input stage, you can choose more complex modulation schemes and get more bits into the zero crossing. maybe something loosely based on IRIG-x14. if your demodulator is fast enough, pulse length could be used to convey data. even DSSS, if you have a stable enough local reference clock

>>1700820
Thanks man. More good news then, CRYSTALS have arrived. time to carefully bend them into place and see if my color timings on the VTR are any better

>>1700948
Just that the clipping is done via normal diodes instead of with sneers, so the signal is clipped to 0.7V.
>resonant tank
I guess I’ll be able to get a somewhat higher Q value than a high-pass, and get rid of any higher frequency harmonics.
>one bit per zero-crossing
Ew, perhaps I’ll throw some error correction algorithm in there? Or modulate my data by frequency-shift keying and lock onto it with a PLL (instead of on-off keying) to avoid that kind of noise. I certainly would prefer to transmit my data on the whole waveform.

For gimmick/learning reasons I’d like to design my own data protocol, and to this end I may want to be able to implement it solely with logic ICs, EEPROM, and simple parallel or serial ADCs and sensors. Throwing a cyclic redundancy check in there would be fun too. I think framing bits and other formatting could be done with a third and maybe forth frequency symbol, otherwise decoding all the formatting would be even more of a nightmare than encoding it in the first place.

>>1700953
well, not so good news. the colors havent changed at all. but the scope is at least showing the correct waveform now on non-aft mode

>>1700969
hmm, so they probably use the input capture on the timers to count pulses to verify that it's a real signal
>higher Q
which also means longer ringing and lower data rate, because you are filtering out the sidebands from the rising and falling edges of the modulating signal
>whole waveform
you won't be saying that when the load hits that switcher just right and wipes out your carrier band. unless you want to try some higher frequencies but now you're basically doing RF

>>1700985
did you adjust the trimmer?
did you adjust the manual FT for best picture?
if so... it's time to start looking at the record color board

>>1700460
So i wasn't able to figure out what is cuasing the leakage so fuck it, i am going to throw out the mosfet and replace it with a latching relay. The fucker won't be bleeding any charge with a physically open circuit.

>>1700220
Shit I thought they were low torque high rpm. Would it be possible to use it for a belt sander or something?

>>1701018
Yes. Google universal motor torque curve

I have an ic which can accept between 3 and 3.6V
And i want to power it from 1.2V batteries, so i use 3 which is 3x1.2 = 3.6V, perfect
Except they are 1.3V when they are charged which is 3*1.3=3.9V = fried ic
so i was thinking, since a diode has a 0.6V drop, would adding a simple diode between the batteries and IC fix the problem?
3.9 - 0.6 = 3.3V which seems like a perfect solution
the current drawn by the ic is just a dozen ma or so

>>1701070

germaniums drop 0.3V while Schottky diodes drop from 0.15 to 0.45.

>>1701071
>germaniums drop 0.3V while Schottky diodes drop from 0.15 to 0.45.
I know.
I am asking if i can trust the diode drop enough to use it as a poor man's voltage regulator?
As in, does the voltage drop change based on drawn current? Since my ic can draw down to 10ma and up to 100ma (in short burts) and i don't want to find out the hard way that for example at such a small current the diode only drops like 0.1V and my priceless IC is now deceased

>>1701073
>Schottky diodes drop from 0.15 to 0.45.
do you mean that as in one diode can drop .15 and another one can drop .45, or as in one diode can jump between that range of drops randomly during operation?

>>1701073

diodes have very flat responses between 10 and 100mA. so, take a measurement and know it wont vary much. a typical germanium cant handle 100mA, tho. you'd have to find a more robust model than the average.

>>1701076

varies by model.

>>1701076

varies by model under typical conditions.

>>1701084

typical VI curve for a B130LAW schottky

>>1700987
yep manual FT has been fiddled to get the best waveform, in fact, the AFT and my manual waveform about match now. the 3.57mhz trimmer does nothing so I suspect its a playback trimmer not a monitor/rec trimmer (it is on the color play bd after all). I probably should be looking at the color rec bd

>>1701087
I see, i have shottky diodes but i think i will slap a generic 1n4xx one there to have a little bit bigger voltage drop just to be on the safe side

>>1701013
go back to the arduino board

>>1701092
as long as your precious still works below 3.0V, go for it
but if it does you shouldn't have >implied 3.0V was the low output limit

Probably dumb question, but what's up with the tiny boxes in all these electronic component drawers/cases I see? Am I really supposed to bend everything into a U in order to fit in them?

>>1701359
>these
show pic(s)

>>1701359

you probably bought drawers for surface-mount but you have through-hole parts.

>>1701091
so something I noticed, I'm getting really high values across the TR201 3SK39 transistor. I should be getting S-1.4 G-0.2 D-1.4 and I;m getting almost 5V across all three.
Apparently NTE221 is a modern equivalent.

>>1701421

could be you fooled with the AGC pot and it's off its proper range, or more likely, the FET is all shorted. power off and measure resistance between all 3 leads.

>>1701421

could be you fooled with the AGC pot and it's off its proper range, or more likely, the FET is all shorted. power off and measure resistance between all 3 leads.

it would be bad news if the FET is dead, coz in that kind of circuit, the FET's characteristics are very critical. it may even be that they put it various FETs of the same type until they got one that had the right action, before soldering it in. not necessarily, but possible.

>>1701434
I don’t think I messed with anything on this board, that’s ACC not AGC which is TR102 just up above but both FETs. Those values check out okay. As for ACC, I can’t get the waveform to show up on my scope. If the voltages are too high it could explain it. I can test it in circuit yes?

>>1701441
>can’t get the waveform to show up on my scope

if you dont see the scope trace at all: since it's a tiny AC waveform riding atop a large DC signal, you gotta set the scope to AC coupling.

if you do see a flat scope trace: there is no signal there coz the FET has killed it.

they may call it ACC (dunno - maybe automatic color control) but it's a classic AGC circuit. you can see it follows the same topology as the one labeled AGC: two caps and diodes on the output of the amp, create a DC voltage that's fed back to the FET on the input to turn on the FET and short-circuit some of the signal to ground to reduce overall gain.

>>1701421
did you get your chinkydinky frequency counter? is the 4.2672MHz on TP204 correct?

>>1701457
ah AC..ill give it a try
>>1701459
i just ordered it Wednesday, i thought i could get by without it but i might as well get it. should be here next week. my belts are STILL with the parts guy so i havent been able to play any tapes but ive been itching to work on this thing, so trying to fix the color circuits is where i'm at.

>>1701469
>>1701457
>AC
that did it, how did you know thats what it needed? anyways, scope is set to 20mv x10 probe and the sheet says 400mv PP so I assume its the middle burst. I adjusted the pot to get it right

>>1701482
>how did you know thats what it needed?

i spend one minute playing with my scope for every hour i spend playing with myself.

>>1701509
Well that works

>>1701509
this is the only correct answer
to everything
ever

>>1701509
I spend more hours with my scope than with my dong these days. It's right next to my computer, so sometimes I just flick it on, plug in the function generator, and start messing with settings. Now I know what holdoff, trigger lock, and the three horizontal modes do. If I ever need to dissect a digital signal, I now know to set my time-per-div to encompass a full data packet, tweak the holdoff until I get a coherent waveform, and flick it to ALT. Then I can set the zoomed-in time-per-div to something finer and zoom about the on-screen wave while the cursors tell me where on the main waveform I'm looking. Pic related.

As I learn more about electricity, the more it scares me. I'm genuinely terrified of electricity now. It's an invisible force that can just instantly kill you. Is this a normal progression?

I have my boyfriend one of my 87V for his birthday and bought a 28II to replace it, it's fucking epic and based.
Feels good to know my puppy has a nice quality, reliable and safe meter in his paws now.

I gave my boyfriend one of my 87V for his birthday and bought a 28II to replace it, it's fucking epic and based.
Feels good to know my puppy has a nice quality, reliable and safe meter in his paws now.

>>1701579
>IP67
I wonder what the seal of that rotary switch is like? Did you take it apart yet?

>>1701584
I haven't taken it apart, this is the first fluke meter I've seen with a tamper seal on it and I dont wanna void my warranty but I think eevblog tore one apart.

hey /ohm/
it's been a while, but to get to the point, i have instrumentation amplifiers ISL28634 and ISL28534
I've read the datasheet quite thoroughly.
I've tried setting the gain set at different gain settings.
The output does change but not by how much i believe i have set the gain to.
What the fuck am I missing?
The only gain setting that i've tested (1, 10, 100, 1000) that is accurate is gain setting of 1.
I am only testing DC signals btw
I'm ouputting small voltages with DC function on my function gen
The voltages at the inputs are accurate to what i have the function gen set at (10mV is actually 10mV)
The output seems to follow:
Gain of 1 is actually gain of 1
Gain of 10 is actually gain of 1.67
Gain of 100 and gain of 1000 is actually a gain of 2

>>1701739

i'm guessing all your instruments are grounded, so when taking measurements, the negative input is being grounded, which is defeating the utility of having differential inputs.

>>1701763
But that shouldn't really matter since i'm measuring a difference from ground too. V- is referenced to ground and the differential stage is also referenced to ground.
It still wouldn't explain why I'm not getting the gain I expect though.

>>1701772
Post circuit, reduce things to basic, make a inverting amplifier or a comparator idk, if that works then go to the next thing until you find what is it you doing wrong

>>1701800
its an ISL28534 instrumentation amplifier. I'm not making an instrumentation amplifier with 3 discrete op amps

>>1701772
sounds like it might not work because it isn't rail-to-rail (enough), try pulling the -ve input to a virtual ground halfway between Vcc and Vee, and outputting your test voltage with respect to the same virtual ground. A simple resistive divider will work well enough for this case, since the input impedances you're dealing with are massive. Though it would require some care to get your power rails isolated from your testing power supply.

>>1701959
Sorry I left out a detail or two
I've been powering it with 4.5V (which is within its operating voltage range)
And using an input voltage between 10mV to 1V (usually playing with 10 - 100mV in 10mV increments and then 100mV to 1V in 100mV increments) And the gain behaviour is the same through the that whole listed range. I've been very careful not to make the output too close to the rails just in case it was that, but nope. same with the inputs.

I'll try that virtual ground thing but I'm almost certain I've done that with the ISL28634 (which i got my hands on first0 and it had similar behaviour too. So i must be doing something wrong. somehow
>get your power rails isolated from your testing power supply.
Why would I need to do that? If you're worried about me shorting something with the common earth ground of my function gen and oscilloscope, i've been careful not to do that either, but I heed your warning anon

>>1701967
>Why would I need to do that?
Just to avoid shorting the common ground of the 10-100mV test source with the instamp's power supply's Vee. If your power supply is split-rail by design then you wouldn't have to worry about that as the 0V between the ±V rails would and should be equal to the ground of the 10-100mV test source.

>>1701575
the next step is respect. don't grab it by the pussy and you'll most likely live

>>1701967
>an input voltage
>an
what about the other one?

>>1698538
Alright, /diy/, this one is kicking my ass and I am embarrassed to even be asking about it because I feel like I'm way over complicating things.

So I got this AC motor on a drill press. It's capable of being able to be reversed by swapping two pairs of wires. This is different than most of the ones I've read online which only require a single pair to be reversed. If that were the case I'd have this solved.

What's kicking my ass is physically translating the circuitry to a type of switch that would allow me to run the motor in forward, reverse, or center the switch for off.

From my thoughts, I basically need a switch with four separate circuits on it, operated by a single lever in a double throw motion. Each circuit would have 3 contacts on it, one from each wire to be reversed, and one to feed back to the motor. Obviously would have to short some contacts together as needed depending on what position the switch is in.

Kicker is that it has to be able to handle wall power which is 120VAC, this particular motor drawing just shy of 7 amps at that voltage.

Does such a switch exist? Is it just called a 4PDT or something?

God this feels like a stupid ass question to be asking but I'd love to be able to run my drill press in reverse.

Alternate solution is just to stick a damn SPST toggle on the hot and be done with it and just flip one end of the drive belt in the seldom chance I need to run it in reverse for some reason....At this point I'm more interested in overcoming my mental block than actually implementing the solution for any benefit. It definitely needs a damn on/off switch though because at the moment it just runs when plugged in to wall power.

>>1701994
> It's capable of being able to be reversed by swapping two pairs of wires.
Can you draw a diagram? Or provide an unambiguous description of what connects to what for forward and reverse operation?

>>1701994
>Is it just called a 4PDT or something?
yes
post diagram of motor windings with reversing information

>>1701997
do you think we'd get a more helpful response by asking anons to post tits?

>>1701996
I have a poorly thought out diagram. It will be more useful to post the plate that covers the wiring connections on the motor as
>>1701997
said.

I'm just not familiar enough with AC motors to know what the other pair would be.

>>1702013
Also here are the lugs that are under that panel on the back of the motor if it matters.

I'm planning on replacing that cord with something more substantial, and actually grounding the motor properly.

>>1701991
see >>1701772
"the other one" (V-) is grounded, since i am measuring (and intend to in its final application) V- referenced voltages

>>1702022
>"the other one" (V-) is grounded
y tho

>>1702029
I've read that too but it doesn't matter because the odd gain behaviour still applies above 100mV input voltage if you read my previous post

>>1702029
>y tho

coz it might work. when what you're doing is failing, you try things. at least, reasonable people do that.

>>1702015

what you need is a 4PDT center off, aka as a ON-OFF-ON. plenty easy to find.

>>1701994
forwards-stop-reverse for tools is called a drum switch

>>1702035
>reasonable people violate the published specifications and whine to other people about their failure as if they were not clearly and obviously failing to hold up their end of the bargain
fuck off back to
>>>/pol/
merchant

>>1702035
>>1702034
Disregard that - I suck cocks
I forgot to keep the V- 100mV (just as you suggested) in every case. I only ever made V+ above 100mV
Thank you anon

>>1702035
>>1702039
Beautiful, this is exactly what I needed.

Thanks anons.

>>1702048
>this is exactly what I needed.

actually it isnt. i changed my mind. instead of a 4PDT ON-OFF-ON i personally would use 2 separate switches: one for direction, one for on/off. why? coz if the machine starts drilling into your wrist, you should be able to stop it with a wild slap at the ON/OFF switch, which would be a PULL-ON, PUSH-OFF type of big switch. (this is better than reversing the drilling of your wrist)

I know, this is a stupid question, and is not /ohm/ related, but...
Can I pop an RCD by shorting N and PE?

>>1702052
That was also the plan, I was going to switch the neutral in regardless of positioning. That isn't hard to do beyond that. I can get it with a normal on/off single pole switch. You guys got me past the bigger roadblock I had.

>>1702066

if ''PE'' is the gay keyword for ground, then the answer is no, coz N and ground are shorted at the fuse box anyway.

>>1702066

if ''PE'' is the gay keyword for ground, then the answer is no, coz N and ground are shorted at the fuse box anyway.

>>1702067

you always switch the live wire, not the neutral, so that once OFF you can ''safely'' play with the wires AFTER the switch without having to go unplug the cord.

>>1702083
Woops, that's what I meant, hot. It's getting late. I was going to research this at least 2 more times before actually building it onto the press. Still haven't figured out exactly how I want to build it all out yet as far as where the switches go and such. I don't think it originally had any switches, or if it did, they were located on the actual motor which has long since been replaced with an...actually..nicer one from an HVAC system.

>>1702088

no switches on the motor probably means it used a foot switch.

I have a question for anyone familiar with video game consoles. Recently my Nintendo Wii (purchased in 2006) encountered issues with playing discs. I opened the whole thing up, dusted it out, cleaned the tech with 98% alcohol, and put it back together. The console will at least recognize Wii discs I put in, but when I choose to play a game, the Wii will either get an error message after loading the initial company logos or outright crash. However, sometimes it will actually boot up the game, but if I have to pause or cross a loading zone, the game will freeze and the Reset button won't work.

What's odd is that Gamecube discs work just fine. I can play any of those and the Wii won't encounter issues. It's only for Wii discs, and no fix-it guide online seems to address this issue. Do I have a lens issue or a motor issue? Can the lens be fixed if Gamecube discs still work, or do I have to purchase a whole new lens/console?

Thanks.

Was looking for some pliers to crimp tubular connectors, to fancy up muh wires. Didn't before realise they crimp with differing shapes.

Is there a crimp shape that I should prefer?

>>1702140
Hexagonal deforms the inside less, so with a harder wire or ferrule it would take less force to use the hexagonal one. But more deformation also likely means a stronger grip. Also the square ones probably fit more positively inside screw terminals, with more area for the screw to contact.
From what I've seen those ones are only usable for ferrules. Connectors like BNCs that also want a hexagonal crimping can't be used with that sort of tools since the end of the connector is larger than the opening of the tool, so you wouldn't be able to get the tool off after using it on both ends of a piece of RG58. So I'm more likely to get one of the sets with two jaws and interchangeable crimping dies, also because they're more flexible in general. For now I'll stick to fucking about with the needlenose pliers and sometimes side cutters.
Still, those are some nice looking crimped ferrules.

>>1702148
>Hexagonal deforms the inside less, so with a harder wire or ferrule it would take less force to use the hexagonal one. But more deformation also likely means a stronger grip. Also the square ones probably fit more positively inside screw terminals, with more area for the screw to contact.
Ah good points, so I'm leaning towards the square ones now. Thanks!
>>1702148
>Still, those are some nice looking crimped ferrules.
Under 20 bucks on AliExpress.

if the faggots looking for ways to use KiCAD without a schematic are still out there, it's your lucky day. install this plugin
https://github.com/xesscorp/WireIt
and create/delete/swap nets from inside pcbnew
or just say fuck it and use gEDA's old pcb app

>>1702152
>Under 20 bucks on AliExpress
Well I'm a cheap-ass, and never use ferrules or screw terminals, so no for me either way.

>>1702161
>use kicad without a schematic
wat
What would you even use it for then

>>1702175
Drawing straight to the PCB

>>1702066
> Can I pop an RCD by shorting N and PE?
Yes. A RCD is triggered by a difference in current between L and N. Normally, what goes in L comes out N so the two are always equal. If you short N and E (on the circuit protected by the RCD), the return current will be split between N and E so not all of it returns through N.

>>1702083
> coz N and ground are shorted at the fuse box anyway.
True, but leads you to the wrong conclusion. The operation of a RCD isn't based upon where the current ends up but upon how it gets there. If the current passing through the live half of the RCD returns other than via the neutral half, the imbalance triggers the RCD.

>>1702197
Also, the earth and neutral are joined on the primary side of the rcd.

>>1702189
Can't you already do that? I mean the bug checker will tell you your nodes are fucked and won't like connecting two unlabelled nodes together, but you can add footprints and silkscreen and all that. For basic testing it's fine. And can't you just import a hand-written netlist? Assuming there's a notepad++ extension for netlists, or else some lightweight little program for it.

Looking at a netlist now, it's split into categories:
>design
file source, date, title block info, other general data
>components
reference id, symbol name, footprint, description, value
>libparts
symbol data?
>libraries
part libraries included
>nets
component reference ids and pin numbers
It's pretty readable in general.

>>1702189
>>1702214
yes you totally can do that. Disable the design rule checker (bug icon) and then you can connect traces wherever. I've done this to make breakout boards.
that WireIt plugin actually looks kinda useful though, for grounding fixturing pins that exist on the footprint but would otherwise require a custom schematic symbol.

>>1702090
Well it's a craftsman drill press, it doesn't have the original motor which would have been a craftsman branded unit for the year this thing is made (circa 50's) - I assume that original motor had a toggle switch on it.

The motor that's on it now has rubber insulators on either end and appears to be from an HVAC system blower. The press was my late grandpa's and that's what he did for a living, was install HVAC systems, so he probably had this unit laying around.

>>1701457
So regarding the ACC, adjusting the pot does change the voltage on those FET leads, so maybe they aren’t fucked up, however when I adjust the waveform to match, the volts are not to spec with the transistor sheet schematic. ACC stands for automatic chroma control however it does not seem to affect the rainbow issues

>>1702290
Oddly though grays show up ok

>>1702197
Okay, now it makes sense, because when I was installing a socket, GFCI popped, and I guess I shorted accidentally neutral and live.
>>1702083
Ground is confusing, because I call neutral a ground sometimes.

>>1702322
you're confusing yourself by oversimplification. happens to the best of us

>>1702253
>grounding fixturing pins
this is a use case I can get behind. mechanical anchors, die paddles, and so on just got quite a bit easier to handle

>>1702290
>not to spec
but is it high or low?
did you diddle C219? did anything happen?

>>1698538
A while back, I got a box of free transformer bell end covers that included 2 transformers and two chokes with it. I want to use one of the inductors in a tube amplifier project, but I have no idea what the specs are on it besides the value which I found out using one of those Chinese multi tester thingies. How could I find out it's current rating?

I've got an arcade PCB that the previous owner cut a 12v trace on for a mod he didn't finish, that I want to revert. What would be the best way to restore this trace? Given the size, could some flux do it, or would I want a conductor pen that can apply silver?
The cut trace is 6, next to the SNK logo on the connector.

>>1702579
flux and a solder blob is fine. flux, a solder blob, and a small square of copper would be better, esp if there's any significant current being carried. might want to take wick to the inward side of it first to make sure all the solder mask is off, so that you can have a clean joint

>>1702579
>the connector
you need to keep it low profile so it doesn't interfere with the connector
clean the crap that's on it off
clean copper on both sides of the cut
lay a piece of copper wire IN the cut crosswise
leave the wire long enough to hold in place
use flux and solder fill
clip the ends of the wire to the width of the trace
cleanup and go

>>1702322
still too high. but let me ask you this,
see >>1701482, the sheet says 400mv PP burst, does that mean the middle burst or the outer burst? each block is 200mv
if middle burst is 400mv, then the transitor values are still too high, im in the 2V range now >>1701421

as for C219, I havent touched it yet but since the freq. counter is on its way I'll just wait, so i dont fuck it up like the C525 I did for the 3.57mhz crsyal, which is probably way out of spec now.
I DID find this, which could help me understand how the hell chroma and ACC and AGC etc works https://www.americanradiohistory.com/hd2/IDX-Service-&-Sales-IDX/Archive-PF-IDX/IDX/70s/Electronic-Servicing-1978-07-OCR-Page-0043.pdf

>>1702616
what's the best way to clean the copper?

>>1702637
diamond file, hobby blade, fine sandpaper on a popsicle stick, whatever is mildly abrasive and works for you

>>1702627
Pic related. burst looks a little bit rounded to me but nearly the right level, maybe still 5-10% hot. you're using the NTSC generator as the source, right?

>>1702640
yep, NTSC is still feeding in. I noticed the ACC form on the play bd looks pretty ugly compared to the rec bd. maybe that just how it is. I fixed the rec ACC level so its right at level now.

>>1702652
also with that, the FET transistor at the BV board is reading s-3.5/g-2.7/d-3.5 when it should be 1.4/0.2/1.4
and the FET transistor at the EV board is reading s-1.7/g-1.0/d-1.7 when it should be 2.6/1.6/2.6

>>1702565
>How could I find out it's current rating?

ask yourself, ''what bad thing would happen if i put too much current through a coil?'' the obvious answer is heat, smoke, and fire. so, pass enough current til it becomes too hot to hold between fingers, and you have your rating.

>>1702704
Basically this, but arguably you could also measure its resistance, estimate a dissipation rate of perhaps 10W (depends on size), and calculate I = √(P/R) in case you don't have a good enough power supply. You'll probably also want to measure the inductance at different frequencies to see where its usable.
Which is actually something I'd quite like to do, now that I think about it. I measure inductance by making a band-pass RCL filter with a known capacitor, and tuning my function generator's sine frequency until it shows a maximum amplitude on my scope. So by doing this with a dozen different capacitors I could build up an inductance-frequency curve.

>>1702579
Looks like someone's already bridged it with solder and cut it again, possibly multiple times. Get that off of there with solder wick and gratuitous flux to get it flat again.
You could just bridge it again with a solder blob taking care to not get it all the way down on to the card edge section of the pin, but that's prone to cracking. The more reliable method is to use a short length of wire or strip of copper across the gap and basically just tack it down on either side with solder.

>>1702652
aaand one more thing. so i've been reading about how chroma works, there's the 3.57mhz, the 4.26mhz, and 600khz, and the ACC, and AGC. you think its something other than the 3.57 crystal that could be causing those color bars issues?

well lads it's finally happened, I'm getting fired from my job for not being functional enough.
2 years in and I completely burned out.
What kind of job should I be looking for, for a burned-out electrical engineer who also has some mechanical skills? At this place they call(ed) me a "rapid prototyping specialist".
gotta be something with low stress. can't do it anymore, I got burned out like hell. Anything in the path of "traditional" engineering I think is going to be too much... I don't mind making less money.

>>1702956
potato main crop and fodder beet harvests are coming up. weigh spuds into bags and sew them up while your brain freewheels.

>>1702956
since you're obviously in antarctica I'm afraid there's nothing else for you. F

>>1702968
I haven't even been c& (yet) and I think potato farming's a step up

>>1702953
there aren't that many color subcarrier-related frequency sources in there. the rainbow on a steady color field is very stable from line to line. that suggests one or both of the color carriers are off. if it's not the playback color oscillator, it could be the record oscillator, OR it could be the automatic gain circuitry around it malfunctioning. no use guessing until the freq counter comes in

>>1702956
consider reevaluating your perspective on office engineering work. you can improve your productivity and reduce stress by decreasing the amount of detail you put into a given task. cut corners on your standardized work until the job fits within schedule. this comes with risk (if you make mistakes), but clearly some risk is better than a certainty of getting fired for under performing. in reality management will know you're doing this and won't care. they decide how much detail goes into your work by controlling how much work is on your plate.

i'd recommend getting another well paid office job and trying it out.

>>1703061
that's kinda depressing.

>>1703067
so is spending a third of your life in a cubicle. it's not your job to fix broken corporate incentive structures.

>>1702995
Right that’s fair, I’ll wait until the frequency counter comes in.

>>1698538
(I know simulation tools are in the sticky.)

Does anyone have a favorite simulator for analog designs? I'm relatively new to electronics and am trying to better understand the operation of switch-mode power supplies. I've tried implementing the basic designs via every-circuit, but the simulations seem to fail in odd ways. I'm not sure if I've messed up or there are weakness in this simulator.

I do not yet have an oscilloscope on my bench, and so I believe simulation is my best bet for experimenting here. Any recommendations are greatly welcome.

>>1703138
pirate multisim. be sure to add grounds to both sides of any isolation. reduce the simulation timestep to be significantly less than your switching period. if you get simulation errors, the first thing to do is to lower the "shunt resistance" parameter, which is basically a hidden resistor from every node to ground that discourages infinite voltage spikes in your shitty idealized simulation model.

>>1703138
I build switchers all the time without a scope. I also use falstad for interactive simulation of basic switchers. I also gargle cocks

>>1703147
Excellent, many thanks.

Does anyone know of an easy on-off premade timer module? What I want to do is basically switch on-off a relay for certain periods of time, like 7 seconds off then 3 seconds on and repeat it until I cut off power.

So i hotglued a relay on top of an arduino to save space and it turns out that sometimes when the relay triggers the arduino resets
i assume this is coz of the contact arcing, as the relay is triggering mains power
is there something i can do to mitigate this like wrapping the relay in tinfoil or something or is the only solution placing the relay farther from the cpu?

>>1703187
I would first disconnect anything mains related to see if it still happens when the relay (de)activates. If so, it's a magnetic effect and other means are needed.

>>1703191
>I would first disconnect anything mains related to see if it still happens when the relay (de)activates
it does not

>If so, it's a magnetic effect and other means are needed.
what do you mean?
one additional thing i tried was adding a snubber across the relay contacts and it helped, but it merely reduced the resetting frequency instead of completely eliminating it

>>1703179

a 555 will do the job. look up the Astable formulas to determine resistor values at https://en.wikipedia.org/wiki/555_timer_IC
e.g.: t(high) = ln(2) x (R1+R2) x C

>>1703193
>>If so, it's a magnetic effect and other means are needed.
>what do you mean?

the energized coil creates a magnetic field which is capable of generating unwanted currents in the arduino. you'd need to neutralize those by moving the coil away (maybe by turning it upside down) or inserting a thick (2-3mm) bit of steel in between.

>>1703220
>you'd need to neutralize those by moving the coil away (maybe by turning it upside down) or inserting a thick (2-3mm) bit of steel in between.
Or rotating the coil 90 degrees, like they've been doing in telephony for over 100 years.

>>1703061
So what you're saying is there's no escape?
Is it time for me to go lick a 13800v wire?

>>1703187
Did you connect the relay coil directly to an Arduino IO pin? Don't do that! The pins aren't meant to supply that much current which can cause a brownout or permanently damage the pin. Additionally when a coil stops getting power the field collapses and creates back EMF that can destroy the pin. Always use a transistor and snubber diode.

>>1703187
1. Place the relay further away
2. Put a snubber network (small cap + series resistor) across the contacts
3. Put a diode reverse biased across the coil
This is a problem even on big industrial machinery. Mechanical relays right next to a brainbox is a no-go. Especially if you're switching inductive loads with it.

>>1703179
look in the industrial controls section of McMaster-Carr/Grainger/Zoro/whatever

>>1703268
if you're feeling enterprising, and you know some people, you could take on odd jobs, a layout here, a circuit there. hop on one of the poo in project freelancer sites and see if it's something worth doing

>>1703187
there's an arduino and microcontrollers general that's probably more on your level

Any tips when searching for parts by their marking?
I was unable to find pic related with any combination of the text.
AllDataSheet, Google, nothing. They are probably Vishay parts, but their site again could not find a match.

>>1703399
if you have a guess at the function (DDR active terminator? ESD protection?) you could try coming at it from the other direction

>>1703417
Right, forgot to mention it's a VRM on the GA-Z170-HD3P motherboard and those two chips are the high- and low-side FETs. Even with this info I had no luck.

>>1700460
Umm is your transistor a mosfet or a jfet but more importantly as drawn you would bec forward biasing the body diode if an N channel mosfet or the gs junction of the jfet which is probably where your current is going and the uc isn’t actually off. If you want a switch that would work like that use a p channel mosfet with a little circuit to latch it (like a flip flop made out of transistors).

Im using an ads1256 and teensy to read a hall effect sensor. The ads1256 is a fast adc (24 bits 30k sps). The issue im having is that the second reading is always a jump higher. For instance if I read and average 1000 samples at a time Ill get data that looks like
>4.6v
>4.4v
>4.6v
>4.4v
Where the second or even numbered samples are always a bit higher. The only thing I could think of is that the voltage source is not clean enough.
Could this be the case the the voltage from the teensy is inconsistent. Would a capacitor help remedy this. My issue is that its always consistently higher so I feel like it could be the code or the ads itself.

>>1703502
If it's not a part of your circuit, I can only assume it's noise that's semi-synchronous with your sample rate. If you're reading at 30k sps it could be a 15kHz noise coming from your power supply. Try FFTing a bunch of the data and see if there are any other multiples or factors of 30kHz. That kind of frequency could conceivably be coming from your power supply, so ensuring you've got plenty of ripple rejection on both the PSU output (e.g. a linreg) and on the input to the ADC (bypass cap). Though admittedly the chance that you happen to have a noise source at half your sample frequency is pretty low. Changing your sample frequency would also do some work towards searching for the source of the issue.
I haven't read the datasheet, but if every second value is trying to sample from a second channel and there's some difference in the internal or external circuitry leading to that other value, that could explain the difference. What are you measuring with your ADC anyhow? A potentiometer?

>>1703523
Measuring the output of a linear halleffect sensor. At first I thought the inconsistency was due to the sensor then I noticed the pattern.

>>1703399
https://www.mouser.ca/datasheet/2/427/sira18adp-553144.pdf
wow so hard anon

I have several broken guitar amps from the 60s to 80s. Some have intermittent static, some only work partially, and some don't make any sound at all. Is an oscilloscope the best way to find where the issue is? Where do I source parts from this era?

>>1703588
>source parts
you don't. you use modern substitutes according to the specs required
>an oscilloscope
but can you look at an amp schematic and understand what's going on and why each component is there and what it's doing, and can you efficiently reverse-engineer whatever amp happens to be in front of you? if not, find a local mentor to walk you through it or don't bother
very often, probing around with a high-impedance high-voltage voltmeter is sufficient to find a problem

>>1703502
>24 bits or 30ksps
fixed
at 30ksps the noise-free resolution is 17.1 bits at PGA=1
>the voltage source is not clean enough
quite possible
>the code
make sure you're using the correct SPI mode, the ADS could be seeing an extra clock pulse and doing something weird
>the ads
not necessarily likely, but poor ground planning could do that
>could a capacitor help
that sounds like something easy and simple enough that you should have tried it already before posting
>?
what happens when you adjust the sample rate down a bit? does the same pattern maintain? is the clock stable?
please don't tell me you're trying to use a high-precision ADC with fucking Dupont leads

Hi. I'm making a LoRa-based project with the Ra-01 module, and I'm trying to make a PCB for it.
Due to size constraints, I'm trying to look for alternatives to the helical antenna provided with the module. I've seen you can use SMD antennas, but those seem to be very sensitive to the ground plane provided, and I don't have any way to test I implemented it optimally.

Given I'm trying to send at either 434MHz or 868MHz, what kind of antenna would you recommend, optimizing for size?

>>1703567
thank fren

>>1698538
Recommend me a soldering iron/station. I've been looking at the TS100. I will probably be using it for SMD and some THT. Something starter that is better than a chinese copy of pic related.

>>1703588
Mr. Carlson's Lab has a video on troubleshooting amplifiers. "Become A Super Star Electronics Troubleshooter" may be a good video to watch before you start poking around.

https://youtu.be/7VBVSEevXaY

>>1703767
I like my TS100. It does everything a soldering iron should plus being very flexible with regards to power. Usually I run it off a battery pack which is nice. but I can't seem to find a chisel tip smaller than 3mm for it. I use a 1.5mm tip on my Weller for almost everything, SMD and THT. The 3mm is kinda too big.

>>1703619
I have a couple voltmeters, I just also have access to an oscilloscope. I know enough about electronics to replace obviously broken things, and build very simple circuits. I don't really understand what everything does though. Luckily my father is still around, and he used to build radios and stuff.
>>1703789
Thanks

>>1703809
I heard it works with some hakko tips. Did you try those?

>>1703729
>project
for your own use or for commercial distribution?
consider just wrapping a 1/4 wavelength trace around the edge of the board

>>1703812
oh good, you do have a mentor.

>>1703767
TS100 is a /g/ meme
get a 936D clone on ali, it doesn't even matter which one, and get a set of bevel tips

>>1703822
>TS100 is a /g/ meme
I'm writing this on a thinkpad from 2012, my desktop runs gentoo, I am wearing a modded f91w, I own 2 raspberry pis and I haven't used windows in 5 years. I AM /g/.
Actually the whole reason I want a new soldering iron is because I want to mod my f91w even more.

>>1703825
then it'll be perfect for you. glhf

>>1703767
>TNI-U
huh

>>1703815
I didn't hear that so I didn't try. But now I'm going to look into it.

>>1703982
it's chinese copies all the way down, brother

>>1704000
not him but speaking of chinese copies i'm using a jabe UD1200 soldering station. can't speak about its temp regulation but it definitely heats up in 2 sec just like real jbc ones

>>1698538
I am a beginner in electronics and I have dumb question that I could not get an answer with googling.
Why do some electronics (like small power supplies, in this case a for a servo) still have - and + terminals for voltage output instead of Vout and Ground? What use is there in + and - polarity in electronics outside making sure regular people put their batteries in the right orientation?

okay big brain time
i have led lights in my house and i want to be able to dim them, but of course leds don't work with dimmers unless you have some special leds which i don't
so what if i connected the leds through an ssr and then switched the ssr with a pwm and changed the brightness of the leds by changing the duty cycle, would that work?

>>1704219
Entirely depends on the circuitry of the LED. Can you measure the current through it? If V*I is significantly larger than the rated power it's likely using a capacitive dropper. If the two values are very similar (i.e. power factor ≈ 1) then it's likely to be a high-LED count linearly regulated model of some kind without filtration. If the power factor is high but not that close to 1, it may be a linear model with capacitive filtration on the input.
For better troubleshooting you'd probably want to look at voltage and current with an oscilloscope, which sounds like dodgy shit to me unless you have an isolation transformer or a mobile scope. Makes me wonder if capacitively coupling probes would be linear enough for these sorts of HV applications, along with the existing magnetic probes (CT/hall) for current.

Dimming at a higher frequency with a SSR would barely be better than a normal phase-fired dimmer in most cases, as IIRC all but the fully-linear models have capacitors in there somewhere that act differently to having the higher-frequency components put through/across them, and it could even be worse in some cases. Got any hints as to what sort of circuitry there might be inside? A bigclive video of a similar model would work.

>>1704239
They are these exact ones, 230V 5W - warm white
https://www.aliexpress.com/item/32850583442.html?spm=a2g0o.cart.0.0.76913c00cgAml5&mp=1

>>1704242
Oh and they are the ones with NO dimmer support

>>1704243
also the PWM freqeuncy wouldn't be super high (since the ssr is a slow piece of shit with 10ms reaction time), so only about 30 or 40hz

>>1704242
>>1704243
> [Voltage] : 220V (Dimmable); AC 85-265V (Not Dimmable)
This means they probably have a switched mode power supply inside them, which will automatically rectify and filter mains electricity to an arbitrary DC value, then buck/flyback convert that to a constant current with which to feed the LEDs. No amount of fucking with that waveform will change the brightness output short of taking the bulb apart and fucking about with the circuitry inside. It might be possible to feed it a very low duty-cycle high frequency wave that RMSs to significantly below 85V to starve the LEDs for current and hope the controller still works enough to keep them on, but I wouldn't recommend doing so.

>>1704245
Then why even suggest it over a phase-fired TRIAC?

>>1704246
>Then why even suggest it over a phase-fired TRIAC?
because i already own an ssr and i also want to control the duty cycle with an IOT device

i guess i will just try it out on one led and see if it works without blowing it up

do you think the leds with dimmable support don't have any fancy power supply in them so they are able to be switched at high frequencies?

>>1704249
>do you think the leds with dimmable support don't have any fancy power supply in them
Since the voltage rating for the dimmable ones was a constant 220V instead of the wide range that SMPSs can handle, probably. But it's also possible that they have a more sophisticated switching control circuit that does everything that I described the non-dimmable controller doing but with an extra monitoring circuit for incoming PWM or dimming waveforms that it converts into a variable current or maybe PWM output. At least I hope that those exist. I can't say it's likely for a chinky bulb like that to be so sophistocated, but considering it only has 3-5 (multi)LED chips in it (assuming each one isn't 50V or 100V), I can't imagine the voltage across them all is high enough to have any sort of linear current regulation, and even a capacitive dropper would likely not work properly on a dimmed waveform. Perhaps look to see if bigclive has reviewed any dimmable bulbs with low LED count.

As far as TRIACs go, there ought to be some circuits that use a PWM output to feed a low-pass filter (a primitive DAC) to vary the threshold of a DIAC along with some protection circuitry.

Where I can find a remote like this, but for lighting? I just need 4 buttons, on/off, +, - and maybe a MODE button.

>>1704274
Get a generic (automotive RF?) remote and put the right labels on it. Then use an IR/RF receiver to feed a MCU on the receiving end. Unless you already have a receiver that you want to get something compatible for, not sure what to do in that case other than fuck about with cool single-PCB designs with an inline CR2015 cell and a bunch of capacitive buttons on the top.

/g/ told me to ask this here

I'm checking out 2 solar panels right now. One is 21W and offers the ability to charge 2 devices simultaneously with 3A output (I'll be using them to charge power banks). The other is 14W, can charge 2 devices simultaneously, but is limited to 2A. They both advertise as 5V so how do they come to 14W and 21W? Shouldn't it be 15W and 10W? W=V*A

Also, a separate question. For maximum efficiency, should I be simultaneously charging 2 powerbanks at the same time, or one at a time? I'm wondering if charging 2 at the same time would cause any problems or be way too slow since 2 powerbanks probably need more than the 2A or 3A that the two solar chargers output

I don't know a lot about this stuff, but I like to learn. Thanks for the help.

I should add the 21W model specifically says 2A while one device is connected and 3A output while 2 are connected. I'm guessing this means each device receives 1.5A? The 14W model just advertises as Maximum 2A output, regardless of 1 or 2 devices plugged in. So if 2 were plugged in each device would receive 1A? Just adding this in case it helps judge efficiency

>>1704299
>Shouldn't it be 15W and 10W? W=V*A
the listed wattage for those panels is for a given amount of sunlight. probably something ridiculous like 1kW/m2. overrating the panels allows you to get the full charging current when less light is available.

>For maximum efficiency, should I be simultaneously charging 2 powerbanks at the same time
charging one will be more efficient. the difference could range from negligible to significant depending on the internal circuitry of your panel.

>So if 2 were plugged in each device would receive 1A?
one would hope, but you can't ever assume with chinkshit. current negotiation over usb is complicated. good odds that neither the panel or the batteries bothered implementing that, so connecting a total maximum system load in excess of the rating could just cause the voltage on your panel to drop out. just connect it up and see what happens.

>>1704299
Note by "2A" USB chargers usually just mean they put the old iPad standard resistors in, most devices will identify that as something they should at least try to pull more than 500mA from, the actual current you'll get will depend on both how much sunlight it's in and how aggressively the phone/power bank/whatever tried to find the maximum current. To figure out what's actually happening a USB power meter would be useful.

>>1704299
>at the same time, or one at a time?

C is the charge (Ah) to be transferred
Case1: one after the other
time = C/2A + C/2A = 2C/2A
Case2: both in parallel
time = 2C/3A

Assumptions are a dime a dozen,
especially after dark..

I have this big old sanding wheel that keeps blowing fuses, I think it's just too much load on the motor. What should I do? I was thinking about maybe some kind of PWM setup that would scale the motor speed up over time

>>1704158
- isn't necessarily ground

>>1704337
For an AC motor, you typically need a variable-frequency drive (VFD). Which may cost more than a new motor.

How much current should it be drawing? What is the rating on the fuse? If it's blowing a fuse which is much higher than the expected current draw, you probably have an intermittent short somewhere.

>>1704096
>UD1200
huh. cute! added to wish list

>>1704249
>I have junk, what do?
look at his pain and laugh

>>1704274
why don't you just get that one and learn to interpret its outputs

>>1704279
automotive key fob transmitters will probably require a matching receiver because you don't want just any roody-poo with a tiny bit of technical skill or a hacked-up garage door opener to gank yo shit and neither do the loan companies
there are HT12D-flavored RF key fob remotes that can be received with the $0.50 RF receiver boards from ali and can be interpreted with a bit of coding, or an HT12E chip if you're not into all that

>>1704299
tell /g/ we don't give a fuck about consumer product shit here and they need to do their fucking job

>>1704337
maybe your start cap or switch is fuxored
you coyly did not follow /ohm/ RULE 2 and state what kind of motor it is, what fuses are blowing (on-device or off), whether the motor is even fit for your sanding wheel or whether you naïvely just mashed shit together, etc. etc. etc.

looks like bomb making is going main stream
back in my day terrorists had enough decency to at least avoid breadboard and arduinos
step it up Hollywood

>>1704384
if you use an arduino, the terrorists have already won

>>1704394
inside job

putting in some oem fogs in my car. 'port' in fuse box is taken up by alarm, need to splice/split the 8awg wire off into a female harness (for fog harness) while retaining the already-there connection going into the fuse box.

Recommendations? C-type splice ? Strange Y splice (that gets clamped)?

Space is really tiny to work in

>>1704453
y u no pic related

>>1704096
>UD1200
>search on ali
>see a female-to-female IEC cable
>see pic related
>220V output socket
https://www.aliexpress.com/item/4000084669385.html
this is not ok

>>1704457
double your power

>>1704456

fog connection with male harness already has one in-line. if i dont connect through fuse box im not getting any power to it. since the cars already wired for it, piggy backing off an empty acc slot in the fuse box might be a waste of time and of a potential slot for another mod connection

>>1704457
seems like the perfect place to plug a fume fan imo

I have only seen datasheets with the data laid out in hex format or something similar before. How would I go about reading this? It's over RS232 btw, and I've tried things like putty and typing the commands in verbatim, but nothing. This is also all they give you for the communication section too.

>>1704486
perfect place to fucking electrocute yourself, power outputs are supposed to use female IECs, the ones without the exposed pins. The ones with exposed pins should only be used for power inputs. Pic related.

>>1704337
clean and relubricate the bearings

>>1704505
they used to name control characters back in the day. HT = horizontal tab = 0x09, EM = end of medium = 0x19, and so on
https://www.ascii-code.com/
>inverted exclamation point instead of lowercase i
kek

>>1704505
Seconding ASCII, I was dealing with a plotter interface that shat out ASCII bytes even for integers. Mine was an HPGL format, perhaps there's a proper name (and hence documentation) for your format too.

>>1704534
there is no standard across printers. "Epson" as a loose pseudo-convention for character printers is about the closest we've gotten in the past 40 years
besides he's got the documentation which is enough. kids these days I swear

>>1704550
>in the past 40 years
Surely printers these days use something far more sophisticated than whatever that ASCII RS232 rubbish is, right? We have USB after all, but I suspect it's all a relatively proprietary serial communication. I wonder how far we are from innately class-compliant driver-free printers that just plug and play. All you'd need is to include a universal printer spec in the USB firmware that ships with every OS / BIOS. Give it plenty of leeway for inkjets, lasers, thermal printers, different colour spaces, different pixel densities, different paper dimensions, etc. Sure it would require more processing on the side of the printer itself, but these days a dozen pages of PDF or jpeg pre-print processing shouldn't be a big issue.

On a side note, I still haven't figured out how to get my inkjet to print at high ink density (i.e. CMYK all at once for maximum opacity) for my etching transparencies, and print to scale at the same time. Printing from my browser lets me do the former but not the latter, printing from inkscape lets me do the latter but not the former.

>>1704556
do notice I said character printers, not graphics printers. if you couldn't infer, anon was working on driving a character-based slip printer
>I wonder how far we are from innately class-compliant driver-free printers that just plug and play.
it's not profitable to enumerate every possible capability of printers extant and yet to be invented, and the drawing models of every printer framework for OSes extant and yet to be invented. manufacturers won't want to waste time and flash on providing a lowest common denominator unless it's HPGL which is already a de facto universal printer language if you don't look very closely
>universal printer spec
what vendor wants that? where's the differentiation? how will they be able to sell you (((services))) like ink subscriptions and lock you into their own products and consumables?
>printing from inkscape
you need to resize your page in Document Properties to do that, iirc. I did it a few times a few years ago

Hello /ohm/
I'm recreating an induction furnace that I found on Youtube that claim to be able to induce 2000W into whatever is put into the coil. I am concerned how to deliver the power and how much might be too much. The MOSFETS of this circuit are rated for 60 Amps and I was thinking of using a server power supply or maybe 2 in a series. Would the current add to eachother and then be distributed through the 4 paths? This is my first electronics-related project so go easy on me

>>1704566
>if you couldn't infer, anon was working on driving a character-based slip printer
I assumed something along those lines, but I thought you were referring to printers in general.
>it's not profitable
>what vendor wants that
Maybe the chinks will start selling open-source printers that actually get kinda good and some semi-reputable company resells these for twice the price and a significant amount of people actually start buying them, forcing competition on all but the most specialist printers? Especially for non-proprietary ink costs. Though that may largely be wishful thinking, since they're more likely to go into the other extreme of highly proprietary botnet wifi enabled IOT printers controllable via some sort of home assistant, but there's still an inch of hope there considering the somewhat substantial diy IOT movement. If only there was a substantial diy power tool battery movement. Perhaps when Milwaukee releases a vape pen.
>if you don't look very closely
Yes I did notice some peculiarities during my own tussle with HPGL. It seems like there's a lot of different ways to do the same thing, and different manufacturers seem to prefer different methods. In my case it was the custom character that was using an unusual implementation, without a substitute character. At least HPGL is pretty easy to decode considering it (and RS232's) framing methods.

>resize your page in Document Properties to do that
I'll give it a shot next time I'm near an inkjet, thanks. But while I'm at uni I could consider buying some of the laser transparencies (not toner transfer because UV-curing solder mask).

>>1704457
what the fuck. I'm truly amazed, surely it cost them more to get that specialty female-female IEC cable than it would have to get the right panel-mount C13 type socket. the mind boggles.

>>1704589
>2 in series
For a higher total voltage? Sure, if you prevent any ground loops. 2 PSUs in parallel is much more dodgy however and I'd be concerned whether they'd fight one another. What's your FETs' maximum voltage? Also what are S1 through S8?

>>1704602
Yeah it's pretty nuts, no way that would pass any sort of regulations. I even bought one of those panel-mount C13 sockets accidentally to slap on a project before realising my mistake, so it's not like they're rare.