/diy/ - Do It Yourself

barrel jacks > USPee

Ok back again with a question on the dead speaker channel. It turns out both leads marked do not send out a signal, specifcally the right channel. how far back do I trace this to find the dead component. I hope its not one of the germanium transistors. Im already borderline selling this for a loss but I would like both channel to be working knowing im this close to figuring it out. The caps marked in green were already replaced

Stupid question but is 20 watts a lot of power?
That would only power a LED light bulb or similarly low energy device wouldn't it?
Could i use that to charge a battery?

>>2489869
one does not repair old electronics for profit, one does it because one likes to do it.

A bit of a noob question, given an n-bit ADC with a full scale voltage Vfs, is the resolution of the ADC defined as
>Vfs / 2^n
or
>Vfs / (2^n - 1)
Our teacher said it's the former but I vaguely remember seeing the later somewhere?

>>2489878
well sure, i've worked on 4 of these stereo consoles before, but its rewarding when you can fix the problems without spending excess money on things that ended up not being a problem at all.

>>2489876
that's effectively in the medium power range in the grand scheme of things. solar panel?

>>2489882
With 2 bit ADC you can represent 4 different values (intervals)
GND ->| 1 | 2 | 3 | 4 | <- VCC
One such interval represents resolution of ADC.

Second equation is used if you design your ADC's transitions differently as explained here
https://electronics.stackexchange.com/questions/258490/analog-digital-converters-lsb-formula

I broke my RFID card, how to fix it?

>>2490004
Soak in acetone and light on fire. Then get a new card.

>>2490004
If it still works you could slather it in Bondo™ and shrink wrap it. lmao

>>2489869
I can't make heads or tails out of that schematic. It looks like T5 and T7 are supposed to be a totem-pole driver, but while their output (emitters) goes down to provide feedback to T1, it doesn't go to the speaker output. Instead the speaker output is connected to T7's collector. Which is objectively one side of a power rail since it's connected to the transformer. The other node going to the speaker output is just an inductor to ground.
Can anyone else understand it?

>>2489882
It's the upper one, but the maximum voltage is >V(0xFF) = Vref * (2^n - 1) / (2^n)
IIRC. Because 0 is included, also because that's how R2R ladders work. I guess whether the Vfs equals the maximum voltage or the reference voltage depends on the datasheet, and while I'd expect it to be the maximum voltage, it does make the math more confusing.

>>2490061
Its a german stereo console from about 1969, made by Grundig. Didnt realize it would be confusing, here's a picture of the board itself if that makes any difference. The speaker out jack is circled for orientation

>>2490055
No, it's not.
Can I remove the plastic and re-connect the wires
?

>>2490082
It's probably (99%) no good.
The coil is tuned for either 13.56 MHz or 125 kHz. Winding your own would be a nightmare. You could theoretically remove the chip, attach a new antenna, sandwich between paper and laminate it. Big pain in the ass no matter what.

>>2490086
Can't re-solder it?

>>2490087
If the coil flexed and lost its enamel then it's finished. Measure the coil resistance first.

>>2489896
Good guess, but I'm looking at turbines.
I'm in a position to buy or build either wind or water turbines, and I'm trying to assess the best option.

It seems clear that I'd be better building several wind turbines out of recycled alternators than buying a proper one.

But the other option is running a pelton wheel turbine of tank water, because I've got a ten meter rise and could get a much higher impulse pressure.

>>2490061
>Can anyone else understand it?

i had the same reaction: output stages make no sense; must be an error in the drawing.
then saw the OP was not well versed in Englisch so i said nothing coz i no longer have the patience to deal with bad drawings and bad Englisch at the same time.

>>2490082
>Can I remove the plastic and re-connect the wires

i once had a fob that stopped working.
i used an older fob from a diff house (same model, diff code) and just transplanted the coil to new one.
worked fine after.
so, yes, it's worth trying to repair the coil and save $50.

I am measuring ripple rejection of LM7812 and it is just horrible. The input ripple is 500mv p-p at 15V and the output ripple is 0 to 5mv (positive swing only). Why is this happening??

>>2490221
>the output ripple is 0 to 5mv

that's trivial.
dont be an autist: ignore it and move on.

>>2490232
>dont be an autist:
i can't. i want to select 1mv/div and see a straight line on the scope. none of those disgusting 5mv spikes every 8.333ms.

>>2490078
Trace the output stage with a DMM continuity mode to figure out what’s actually connected to what. The schematic is probably nonsense. While you’re there, do diode checks of the BE and BC junctions of the two big output transistors, that usually tells you if they’re dead.

>>2490210
If he’s German it’s probably fine.

>>2490221
What frequency? If it’s not 100-120Hz you’re doing it wrong.

>>2490262
Make a Sziklai pair emitter follower, with a TL431 on its base such that it samples the divided output voltage. Put a big capacitor in parallel with the TL431, and a resistor from Vin to the base. It’s like a zener-BJT regulator but with better regulation. Put a choke before it if you’ve got particularly high frequency ripple.

>>2490210
I AM not german, the stereo is lol. Im full american. Grundig made imports into the US for a long time, its just that the schematics are multilingual. This one actually came with the unit itself which is nice, i didnt have to purchase a SAMS copy.


>>2490269
The way the AC connectors work this amp unit needs to be connected to the body of the console for it to be powered, right now its on my workbench. its it worth it to cut a plug off an extension cord so I can power this? Or can it all be checked passively?

>>2490282
>Or can it all be checked passively?

step 1 is visual, step 2 is passive measurements, step 3 is powered measurements.
visually i see:
red circle = crappy looking solder joints, one of which seems to be missing a pad
blue circle = screws appear very grungy, possibly too oxidized to make contact
yellow circle = PCB is burnt. when it gets burnt badly, it can actually become conductive, so dig into it with ohm-meter

>>2490269
>What frequency?
Solid 120Hz in sync with the ripple on input the capacitor. Attaching an output cap does absolutely nothing.

>>2490276
it is 120Hz so it is expected it is just it is far from 60dB according to the specs.
But maybe this anon is right: >>2490232
I just re-measured and the ripple is all over the place from 2mv to 20mv based on where I attach the ground clip even though it is supposed to be the same potential. I have a bunch of wires twisted together and there is 1amp flowing through them (but before the regulator) and plus I have a cheap scope so who knows what I am really measuring.

>>2490297
Maybe the 60dB is contingent on high dropout? 3V could be higher. Even 60dB is kinda ass. Look into the linear regulator circuits used by audiophiles.

>>2490293
good eye but
red circle - that was after i reheated the cold solder point, it was flat silver and possibly a cold joint. it secreted a lot of black liquid at the base (old flux)? pad is still there. just to be safe i pulled off the old solder and going to add new stuff.
blue = good call, thats the collector pin for the transistor so those will get cleaned
yellow - shadow of the image, nothing on the board is burnt.

>>2489838
Need your help ohm. I’m converting an old 5 bay scsi case to a frrenas/synology NAS. I replaced the board with a tiny little SOC board (dual celeron and 4gb ram). The old enclosure took IDE drives so I’ve cobbled together drop in replacements that take sata instead.

I’m almost done soldering, and think I should add some caps on the 5v and 12v to ground. What size caps do i use? Can I use regular can electrolytic?

>>2490319
Before I found the 5 bay enclosure, I tore apart a 2 bay sata one that had this board in it, which has the two sized caps on 12v and 5v.

I guess I can unsolder them and try to see if my eBay component tester is accurate enough

>>2490321
1uF and 10uF probably

>>2490319
Regular should be fine, but low-ESR 105C would be my desire. Just to ensure it’s rated well. Also overshoot the voltage ratings for longevity’s sake.

I think it is probably me.
I get close to zero ripple when I measure right at the regulator Out/Gnd pins. But the farther I connect the ground clip from the regulator ground, the more ripple I get. Is that due to a ground loop? There is a high current flowing through R1. So if I connect ground to the left of R1, especially right on the bridge, I measure a lot of ripple. Is that expected?

>>2490333
Is there an actual protective earth connection somewhere in the circuit? Or just through the scope probe? What's the load current out of the regulator? IIRC they're not designed for good ripple rejection at no load.

>>2490323
>>2490324
Thanks. So I have enough 1uf (top left) and 10uf (top right), but they are “eBay” chinkshit.

>>2490339
I have some good quality ones i pulled off various boards.

>>2490339
Might be fine if you put ceramics or film caps in parallel.

>>2490338
No, no protective earth, just the probe.
I tried both at no load and at 10mA and there is no detectable difference on the scope.

>IIRC they're not designed for good ripple rejection at no load.
Could be but technically they are never truly idle since there is an internal voltage divider connected across the output that sets V- for the error amplifier. The bias current according to the datasheet is about 4-8mA.

>>2490323
>>2490324
>>2490341
>>2490319
I don’t have hdds yet, so I ’ll just buy the correct kind and wait. Can you link me to the mouser or (non China) eBay ones that are lowest/best? (Sorry I’m kinda a newbie to EE, I’m a mole bio benchie by trade)

>>2490347
Yeah but they usually want more like 100mA load.

are laptop mobos worth trying to fix? Got a laptop that doesn't turn on and as soon as I plug in the power I hear a small click as if theres a short and the 20v wall power supply suddenly reads like 0.3v until I unplug and replug it. I might have randomly taken off small capacitors here and there so what are the odds I ultrafucked the whole thing?

>>2490380
Do you have a multimeter? You can try measuring for continuity across the power pins, if it’s a short circuit even after waiting for the caps to fill, then you could use a milliohm meter to trace where the problem is. If it isn’t a short it will be difficult to trace. A thermal camera may help if you can take an image of the board right after the PSU cuts off.

Also ensure the power adapter isn’t putting out something stupid like 30V.

>>2490319
>>2490321
Managed to get one off the related board type, my thingy says they are 24ish uF

>>2490380
>are laptop mobos worth trying to fix?
Short answer is yes.
Longer answer is that fixing mobos is helluva hard.

>>2490380
>I might have randomly taken off small capacitors here and there
Kek, why?

Any recommended isolated switching topologies for extremely low noise? Pretty low power, just a watt or three. Gonna wind my own toroidal transformer for low common-mode noise, so I just need to worry about normal-mode noise.

>>2490478
I checked for capacitors that had continuity across them and if they did, I soldered them off. After a while I had about 20-30 of those super tiny ones taken off.

Are caps that important in a laptop motherboard?

>>2490536
Pretty important considering the high frequencies they operate at.

I want to amplify a signal from an AD9833 signal generator. I thought I could by connecting the OUT pin of the ad9833 to the SIG pin of a mosfet module.
Both devices work but the AD9833 has no effect on the mosfet. Should this be possible?

>>2490638
>to the SIG pin of a mosfet module.
Which module? And how much power at what bandwidth?

>>2490638
>the AD9833 has no effect on the mosfet.

seems you're trying to use a logic-level FET to amplify an analog voltage.
that's completely insane.
might even be illegal in some states, so dont answer the door for the next month coz it could be police.

im trying to figure out the amp draw of a light i have and when i set the dvom to the 10A setting and connect the light it arcs
if i use the mA setting i read 0
why?

>>2490763
nvm im dumb

>>2490638

>SIG pin of a mosfet module


dude, the AD9833 has a maximum Vout of like 600 - 650 mV, even at maximum output you're not even close to the threshold of most small signal modules. small signal MOSFET modules especially are designed exclusively for full switching from low level inputs, and have extra on-die features to make that happen, so trying to jockey one in the linear region is a no-go.

This is literally why they make power op-amps, just use one.

>>2490763
Measuring current in parallel not series? RIP mA fuse either way.

>>2490845
>Measuring current in parallel not series?
yeah lol
no multimeters were harmed in this post (they still work somehow)

Has anyone ever tried data transmission over radio frequencies? Is it feasible?

I want to build a time tracker for 20 yard and 40 yard dashs. I want to keep it as analog as possible...

>>2490885
I mean yes I've done radio, but >analog isn't exactly the easiest thing. There's dozens of easy to use radio transceivers.

>>2490898
*easy to use *digital* radio transceivers.

>>2490885
If you want to do it without an ADC before the transmission, then a simple FM audio transmitter IC can send an analogue signal. But I’d make that signal be a square wave switching between the analogue voltage and 0V at 1kHz or so, such that the FM tuning doesn’t just adapt to he new baseline frequency. Then I’d AC-pass that after the demodulator IC and use a diode detector to extract the amplitude. FM means the distance shouldn’t impact the output value.

Could use CD4046s for modulation and demodulation, with the help of a quick op-amp, a BJT power amp, and LC filters. Puts the frequency around the 1MHz mark though, so you’d need to think about your antenna topology.

>>2490885
bluetoof stopwatch? trigger the stop with a IR sensor/emitter at the finish line.

Turns out there's software called UVtools that will take a PCB gerber file and output a variety of different SLA printer formats for photoexposing. Neat. But you must align your board up near 0,0, and don't forget to go into the settings and set it to the right printer.

Anyhow, after an hour or two of working it out without any documentation, it's giving me square and to-scale results, and should be a nice and simple workflow going forwards. Shame I can't increase the exposure times beyond 200s though, I don't think my replacement backlight driver board is good enough for that. Especially when doing solder mask.

anyone have any experience with Fachkunde Elektrotechnik by Klaus Tkotz? it's translation seems to be only decently reviewed book in my country (not like there are many books to chose from)

When I use large capacitors as a toothpick, the leads always taste sour, probably because they never fully discharge. This is a good demonstration of the laws of physics. That exponential decay that never goes to zero. Thanks for reading my blog.

>>2491392
>eating electronic components again
Hi, Ben Honks. When is your Apple Two Sea video going to be uploaded?

>>2491392
I sure hope they didn't have contact with lead.

>>2491398
lead is overrated 2bh. wife is always like omg omg i smell something are you soldering again?? we are all gonna get cancer aaaaaa

>>2491403
It smells like a pine forest fire. It's also why you never get bothered by raccoons or deer when soldering.

>pinecil
I didn't expect it to be so tiny, it's smaller than my dick lmao
Very cute tho, now I just need to mod the firmware to put chen on the idle menu

>>2491407
peencel

So in wireless charging, what happens is that the receiving coil, through the electromagnatic force produced by the tramitter coil, makes the electrons in the battery go from the + side back to the - side?

>>2491436
I have no idea how wireless charging works in your phone.
But if you have the receiver coil produce AC current, then rectify that, presumably into some charging circuitry that prevents spikes and whatever, it should be able to charge a battery just fine.
And electrons go from - -> +

>>2491436
inductive coupling i.e. air-core transformer?
primary coil is the transmitter
secondary coil is the receiver

>>2491445
That wasn't the point tho, i was talking about how the EMF received by the secondary coil is able to make a battery charge.

>>2491452
>inductive coupling i.e. air-core transformer?
yes

>>2491461
When the metal coil is inside a magnetic field that is changing, a current is induced.
Not much more to it than that.

>>2491462
But that current will be in the oposite polarity that the circuit normaly operates in, right?

>>2491465
No it will be alternating current that is rectified at the endpoints using a full bridge rectifier

Wait you mean at the battery?
Yes, applying a voltage in that direction is how you charge a battery.

>>2491467
>Wait you mean at the battery?
Yes. Thanks.

I've got a raspi 4 sitting in my cupboard gathering dust because I overestimated how much I knew about electronics. Now I want to make it into a UMPC like pic related, what parts do I need to achieve this?

>>2491578
More of a /mcg/ question. I’d say the biggest decisions to make will be whether the keyboard and mouse are custom or an off-the-shelf part that the rest is going to be based around. Because you’re going to have to choose a screen to fit everything around anyhow. Gonna need some sort of 3D CNC/printer for the mechanical side.

>>2491578
>what parts do I need

i'd sell the raspi 4, buy a raspi 400, and scotch tape a small HDMI monitor to the back.
(just kidding: no scotch tape, but hinges to make it like a laptop)

>>2491616
I was planning on buying a 3d printer anyway, but I'm sure I could cobble something together as a temporary prototype case. How hard is it to make a custom keyboard?

>>2491647
The 400 doesn't have batteries, though. Power management for the screen and raspi board would probably be the most difficult part of making this work.

>>2491673
Ask the guys over at /g/mkg/. There are simpler ways, like with cheap tactile switches or even capacitive touch pads, but they’re compromises.

As for the 400, you can probably fit a battery in there, maybe with a spacer to make the case wider, but I suspect the keyboard is a bit too large.

Go for seperate half- keyboards on each upper arm, so you type while your arms are crossed. Also use a glasses-based display.

Zack Friedman has a few similar projects.

I was doing deep clean on my razor huntsman and I guess one of the optical components sheared off the board and took a pad with it. I am repairing the pad now and in the meantime goong ro attempt to move one from a rarely used button to this location because I'll never find this tiny ass thing. Where can I get more of these components? Would be nice to restore the whole thing to full working order at some point.

>>2491699

How do you program a drum sequencer in arduino?

>>2491701

>>2491690
>Go for seperate half- keyboards on each upper arm, so you type while your arms are crossed.
That's brilliant but also incredibly autistic. Wearing a wire from one keyboard to the other would be a pain though.

>>2491701
Learn to use hardware timers. The rest is basically trivial, just store a byte or two in RAM for each pulse in the sequence, where each bit represents an instrument being played or not played. Have a rotary encoder to set the number of total pulses, another to set the speed, and maybe have a modifier key to say a press only triggers every second time around. Just have a fixed high-number of pulses per second if you don’t want it all in lockstep. Also channel enable switches.

As for the synthesis of sound, it’s probably best just to keep lo-fi samples stored in progmem and spit them out into a DAC as needed. External memory may help.

complete moron here.
what would be the best way to create a pirate radio station.

and yes pirate radio, not just an online station.
like an actual pirate radio that can be broadcast onto FM or AM.

>>2491773
Class-C or RF Class-D amplifiers aren't particularly difficult to make, but at 100MHz or so you do need somewhat specialised components. Broadcasting AM at only ~1MHz or so is significantly easier, you could use off-the-shelf FETs and gate driver ICs. A class-C/D AM transmitter can be modulated by having your power rail itself be varied about some DC level using an SMPS that allows for an external reference voltage (or some types of class-D audio amplifier). 10-100W or so should be quite doable, though I can't speak for audio fidelity.

The real stealth method is to spit out digital signals at such low frequencies that a license isn't necessary. If you make it narrowband enough, your receivers may get a decently high SNR just by using a very sharp input filter.

>>2490909
Bluetooth would be overkill imo
I really only need a to send a short pulse coming from the IR barrier to the receiver to stop the measurement
But I have never messed around with radio stuff before...

>>2491789
Also most Bluetooth units like the hc05 only have a range of 10 yards

>>2491789
I don't actually think sending out radio signals is legal in my country
What would be an alternative for wireless transmission for up to 100m?

>>2491799
Every country has some sort of ISM (industry scientific medical) band that anyone can use. That's where all your wireless doorbells, garage door openers, WiFi, RC cars/planes and other things operate. Even North Korea probably has this unlicensed band that anyone can use.

>>2489869
you're asking qualified people to do a job for you for free

why are you doing this

>>2491799
>alternative for wireless transmission for up to 100m?

never used it, but keep hearing about Lora technology with range of several kilometers.

do have use a bunch of cheapo FRS walkies. often advertised as 20km range or so but it's more like 400m in the city.

and ali has 8-watt Baofeng walkies at ridiculous prices: https://www.aliexpress.com/w/wholesale-5w-radio.html

I have an older multimeter (hp 3468a) which won't accept test leads with shielded plugs. Has anyone ever changed the jacks on one of these to accept them or made some sort of adapter?
I understand that I could technically just cut the shielding off the lead plugs, but I would rather keep it on so that the protection is there for my other meters which have them

>>2491957
An adapter should be easy to make but best place to post that sort of thing on 4chan is /g/ I would look into amateur radio resources since old hams are often electronics techs and someone else has surely solved that problem.

Raspi and arduino are the most overrated garbage ever.

>>2492037
arduinos are literally just devboards

How do you swedish /ohm/ers deal with the fact that leaded solder is illegal?

>>2491407
Anon, whoever was in the room when you were holding the pinecil they werent wondering if you dick was bigger or smaller than it. No one here was either.

>>2492087
I buy it at my local store anyhow. Then realise the flux they put in it is ass and buy some from LCSC instead.

>>2491700
Well despite your non-answer the repair worked.

>>2492087
must suck living in a nanny-state

>>2492125
It was an answer and I'm glad it helped you, Anon. You get a scratch & sniff banana sticker.

i'd murder the chip manufacturers who make 8-pin opamps with a non-standard pinout. especially the posers who add Voffset trim pins and mess up the standard arrangement out - + gnd

>>2492279
but seriously, either one is fine, as long there are only two of them to remember. but are there any parts that have a different pinout than these two?

After some more testing I found that the mosfet module does work as a switch if I use a 9V battery to switch on and off a 15V battery. It's kinda baffling to me since according to the datasheet 'gate threshold voltage' min is 2V and max is 4V. I guess the ad9833 cant supply enough current to make it work.

>>2490660
Its an IRF520. I dont know what you mean by bandwidth, I'm new at this sorry.
>>2490695
I don't see why it would be illegal, I'm trying to build an echosounder and I will be sure to keep the noise level low enough to not harm marine wildlife,

>>2490780
>so trying to jockey one in the linear region is a no-go.
The above is incomprehensible to me. However, the NJM4556A power op-amp might be what I am looking for. Would that be able to create, say, a 200kHz square wave signal +/-25V? I will connect it to a piezo transducer.

This is the power op amp I mean
https://www.mouser.com/datasheet/2/294/NJM4556A_E-1917507.pdf

>>2492373
If you just want to make a square wave, then you want a half-bridge, maybe an H-bridge. Look into driver ICs like the IR2184 (pic related), or for lower power ratings an integrated solution like the old L298N. Any motor driver chip will be able to output a square wave, and should be able to do so reasonably fast. The AD9833 outputting a square wave into the motor driver or half-bridge controller would do a reasonable job, though I don't see why you need to use an AD9833 as opposed to a simpler oscillator. It would be even simpler in some ways to just make a simple class-B amp with a few power BJTs, if you only need a few watts of output.
A resonant drive can give a lot more net power to the piezo (it's what they do for ultrasonic cleaners and cutters IIRC), but the frequency may drift a bit and it needs a more complicated driving circuit.

Just using a single transistor means you can sink current from the piezo, but nothing will actually pull current through it in the opposite direction.

No need to use a linear power amplifier for a square wave, you'll be slew-rate limited anyhow.

>>2492373
>I guess the ad9833 cant supply enough current to make it work.
Read the datasheet
>SIGNAL DAC SPECIFICATIONS
>VOUT Maximum 0.65 V

> Would that be able to create, say, a 200kHz square wave signal +/-25V? I will connect it to a piezo transducer.
No, because again it says maximum supply is +/-18V so it's impossible to have more output voltage than that without destroying the amplifier. It also depends on what kind of transducer you have. It should have a power rating which will tell you what kind of an amplifier or driver you need.

Before you start doing anything, make sure you understand the basics of whatever device you're trying to use. There are resources in the OP and considering your questions you need to start with fundamentals of voltage, current, resistance/impedance, power and how they're all related. Reading through "Mims III, Getting Started in Electronics" would give you enough knowledge to build what you want.

>>2492379
I'll have to go through all you wrote, it'll take some time as I'm a beginner.
>though I don't see why you need to use an AD9833 as opposed to a simpler oscillator
The idea is to make a sound then wait for the echo, using the time for the echo to arrive to deduce the distance. My idea is to use the interrupt pins on an MCU to measure the time between the start/end of the output sound pulse and the start/end of the (amplified) echo. The AD9833 can communicate with an MCU using spi.
The MCU can do other things also like talk to a computer via usb and measure the water temperature via some sensor (this is needed as speed of sound in water varies based on temperature).

>>2492387
For the half-bridge vs full-bridge, a half-bridge like my pic will give you VDD as the peak-to-peak amplitude, so you'd need 50VDC there for ±25V. But if you used a full-bridge (replace the big capacitor with another half-bridge doing the opposite of the other one) then you'd get twice that; you'd just need VDD to equal 25V.

>SPI
Or you could just spit out a 50% duty-cycle PWM square wave from the MCU, or even bit-bang it. Much less latency that way, assuming you have a hardware timer to spare.
As for the interrupts, you'll need a sensitive enough input amplifier and a comparator of some kind. Will you try to use the piezo itself as the receiver as well as the transmitter?

Also look into time-to-digital converter ICs, they're kinda neat.

If I want to measure a small-ish angle for a self-balancing robot, I couldn't use a MEMS gyro could I? They spit out degrees per second, not degrees. Inertial sounds inaccurate for long-term use. But since the alternative is an accelerometer, it would need to be sensitive to acceleration due to gravity without being swamped by vibrations. Maybe mount it separately from everything else using a damped mechanical system? I'm reading the datasheet of https://www.farnell.com/datasheets/2293691.pdf, and while it alludes to an internal low-pass-filter, I think it's just talking about debouncing of the portrait-to-landscape threshold.

>>2491822
what are you talking about, most of what gets posted here is asking for help on making and diagnosing circuits

>>2492428
>what are you talking about

Ignore what I said. I'm just a money-grubbing Jew who cant help but think in terms of exploiting others.

>>2492428
>>2492430

>>2492392
to detect tilt angle you would typically run something like a complementary filter, Kalman filter or a Madgwick filter combining acceleration and angular rate. You need to get something like an mpu6080 and fuse the data on an mcu at a fast rate (data input at 100Hz at least).
These cheap mems intertial sensors typically have an offset caused by variations in temperature so they need to be calibrated everytime you use them (like a drone needs to be flat and motionless before each flight). Or spend more money on an inertial motion unit that is heated to a constant temperature. Or you build ring laser gyros.

>>2492438
I'd want to detect changes probably no smaller than 50ms. I imagine the gyro is to get inertial measurements of the higher frequencies, but would it be necessary for my case? Not sure what to do about thermal drift, might try to give it some airflow.

Guess I'll get some LSM6DS_ ICs, they look fast enough and do both accel and gyro. If I don't need the gyro I can just not sample it. As for the math side it's mostly over my head for now. Do you think it could be doable with an 8-bit 16MHz micro, assuming 100 samples per second? A maximum of 16-bit precision would probably be fine for all the mathematics.

i need a dummy load that will draw 1 amp from a 9 volt power supply. ideally it would be a fairly sharp current draw, because i'm trying to test the load response of my SEPIC converter.

can this be done with 12v automotive lightbulbs, like SYLVANIA 1157? i keep reading conflicting things, that lightbulbs can be a pain in the ass because they have comparatively low resistance when cold. i'm not sure what this means in the real world -- can i put 4 of these bulbs (rated at 12V, 26W ==> ~2.1R) in series and expect ~1A of draw? or is real world performance going to be a large inrush current that trips the over-current protection on my supply?

(i feel like a retard for asking this. i would just try it myself, but i don't have any bulbs on hand. figured i would ask before wasting the money on something, if it was doomed to failure.)

>>2492587
Measure the cold resistance before you try, you may well need a lot more in series.

For example, I've got a mains bulb that's 18.6Ω, rated at 250W 240VAC. So the AC current while it's hot will be 250/240 = 1.04A, while the cold current will be 240/18.6 = 12.9A. That's a really big jump, a resistance change by a factor of 12.4, which some reading suggests is typical.

Use a 10W 9-10Ω power resistor instead.

>buy a KSger T12 soldering iron
>want to make a Power supply with RD6006
>read about minor grounding issue
>need to use a soldering iron to fix issue
>order a pinecil
>don't have a good idea for a project after the power supply
Oh yeah
It's time to spend some money.
I'm just lucky I don't know enough to use a oscilloscope

>>2492653
>don't have a good idea for a project

here's 3 free ideas
- up-skirt camera
- travel clock
- alarm clock that sounds like a dog puking - ''Because nothing wakes me up and gets me out of bed faster than that.'' - Paul Feig

When they call leds "water clear" it's just a bad translation from chinese isn't it?

>>2492792
it might be a reference to resin, where it may mean something like "colourless AND not diffuse"

>>2492793
no angry words in my general thank you

>>2492535
If you want to make a balancing robot and want to know what angle it tilts at, you will get nowhere by trying to integrate angular rate (degrees per second) from a cheap gyro. I have seen three separate cases in two universities (two Masters thesises and a Phd attempt) of people failing because they don't understand how noisy these sensors are.

Take a look at https://github.com/kriswiner/MPU9250
This is an example of how you can get the pitch angle using a cheap sensor. The mathematics takes a long time to understand so don't worry about it. Get an arduino or similar board, connect an mpu6080 or mpu9250, calibrate them (you can find code for that online) and use the ready-made algorithms to get a reliable pitch angle. I don't know if an 8bit, 16MHz micro will work. When I did a similar project years ago I did the sensible thing by copying people online who already had working setups. Of course I tried to do the whole 'integrating acceleration twice' thing. What I found was that you cannot easily differentiate acceleration due to gravity from acceleration due to motion.

A 'complementary filter' makes two separate calculations: 1) it calculates the roll and pitch from acceleration by assuming the sensors are stationary, 2) it calculates roll and pitch by taking the previous roll and pitch and adding the change of angle obtained by integrating the angular rate. The new roll and pitch estimate is then a combination of both (eg 40% from acceleration and 60% from angular rate). This is a simplification but that's basically how it works.

>>2492806
>What I found was that you cannot easily differentiate acceleration due to gravity from acceleration due to motion.
But if all you want to know is orientation, can't you just use accelerometer data and ignore gyro data? Or are those accelerometers as stupidly noisy as the gyros?

Problem with those MPUs is they're permanently out of stock. I can get the 6040 from LCSC, but for this order I was wanting to order from Element14 instead.
Is there a capability difference between those MPUs and the combination gyro+accel ICs like the LSM6DS? Because reconstructing that 9250 code with any old sensor should be doable.

>>2492824
>But if all you want to know is orientation, can't you just use accelerometer data and ignore gyro data?
If your robot is static, yes. However, imagine your robot starts moving forward without tilting, the acceleration on the x axis will increase so if you are calculating tilt based on acceleration, the algorythm will think your robot is tilting. If you add gyro data, then the gyros will know that the robot didn't rotate. So you get most of the information from the gyros (which drift, but slowly) but gyros alone cant know which way the robot is tilting. Therefore, you add acceleration data (which includes false information but only in the short term since your robot doesn't accelerate for a long time).

>Because reconstructing that 9250 code with any old sensor should be doable.
yes, that's correct. desu I would look for a sensor with LSM6DS with a magnetometer so you can know the yaw but LSM6DS seems to be modern equivalent to mpu6040.

>>2492793
what are you upset about

>>2492087
I get mine from work. Sometimes you can buy from AliExpress or something. Maybe call some smaller electronics company and ask if you can get wome from them

>>2492587
>>2492590
do you have any experience with nichrome? i figure it might be cheaper than building a power resistor bank, with heatsinks and fans and everything.

i found an online calculator that suggests 9v through 36 inches of 26 awg nichrome should pull 1.1 amps and reach 114 C, which isn't too bad of a temperature to deal with. i'm just not sure how i could attach my power supply to it?

>>2492917
>how i could attach my power supply to it?
Wrap a coil around fiberglass cloth or ceramic tube like a vape pen.

>>2492087
Is it actually illegal, or is it just not allowed to sell assembled products that contain lead? In my EU country it's the latter. I can still buy lead solder, but any devices that are sold are guaranteed to be free of lead.

I've been trying a bunch of opamps for my dummy load, and so far one of the best has been COS8551TRA, which is an Asian brand but it is a truly rail-to-rail high precision chopper with an excellent phase margin so it is unity gain stable without any compensation. The only problem is it is 5v. So it cannot drive the MOSFET above 5v and so the maximum current is limited to about 3A. I wonder if I could put some voltage shifter in between the opamp output and the gate. Would a bjt work? I tried some configuration and couldn't get it to work. And the feedback loop would probably become too unpredictable.

Is there a "darlington" mix of BJT driving MOSFET?

something like a push pull maybe?

>>2492988
>The only problem is it is 5v. So it cannot drive the MOSFET above 5v
Or can it?
https://youtu.be/TN25ghkfgQA?t=2
https://www.edn.com/bootstrapping-your-op-amp-yields-wide-voltage-swings/

>>2490078
Okay made some progress but still need some help,
I’m reading 0V across the right channel where there should be 18V, marked in red. So it could be a bad diode. Interesting that on the schematic they’re listed as the same part # for d2 d3, but on the board they are two different pieces.
My question is how do I measure the area on the right, what are the black bars, transformer? I want to see how far back the problem is

>>2493106
Here are the two diode, the top one is the right channel the bottom one is working normally. How can I determine what the spec is on the diode to replace it if needed?

>>2493109
Sorry the picture changed, the one on the left is the bad one

>>2493106
you can use any 1n400_ series diode that will fit through the holes
it looks like someone replaced that diode before

>>2493112
Thank you. A 1n4003 is rated at 200V which should be plenty. I’ll order one and try it out later this week

>>2493122
>I’ll order one
just one? heh

does polarity of opamp inputs matter?
if i'm just using the op amp to amplify AC signal it shouldn't really matter right? At most it will invert the sine on the output which is fine since it's mirrored anyway

>>2493132
Depending on where you feed the signal you'll have either inverting or non-inverting amplifier. But do keep in mind that feedback must always be negative if you want amplification. This means that feedback path from the output must go to inverting (-) input, regardless of where you feed in your signal.

>>2492848
Ah, I think you’re right. It might be doable to seperate acceleration from the constant magnitude of gravity if they’re always normal to one another, but of course acceleration up or down a hill or dropping down an obstacle will make an infinite number of accelerations that can add to gravity to make a given vector.
Yaw isn’t important to me at all.

>>2492917
Either wrap it around something like a brick with notches cut corners, or immerse it in water on a frame made of something. I made a wooden frame to hold 60Ω of the stuff in a saucepan full of water, where I can easily clip onto it to get a desired resistance. Even out of the water it works fine for 10W or so, very little smoking.

As for connections, use alligator clips temporarily or crimped connectors (with silicone or PTFE insulated wire) permanently.

>>2492988
Why do you need RRIO? Any old ground-sensing op-amp (one that uses PNP input transistors to ground) should work fine, and go to higher voltages. There are a few RRIO op-amps that go up to 15V or so.
Or just add a class-A amplifier after it.

>>2493106
Measure AC voltage directly across that winding. Each black bar is a transformer winding.

>>2493132
Post circuit

>>2493122
>A 1n4003 is rated at 200V which should be plenty.

dont think so.
the 2500uF cap suggests the circuit might draw around 2A, whereas the 1N400x series are all rated 1A.
anyway, before you go changing anything, put your meter on AC volts and measure across white/yellow wires.
the black bars are transformer secondaries and will prob read around 15Vac if they're not burnt out.

Electronics noob here, top channel is a 12v rail in a CRT which I want to smooth out. Any suggestions, before I start soldering these random aliexpress ceramic disk capacitors on it?

>>2493144
>Any suggestions

yes: keep at least one eye open.
since you have a scope, you can instantly see the results of any actions you take.

>>2493144
Got a schematic? If so find the section that describes the 12V supply. Depending on what it is you may be able to add a ripple rejection stage before the output regulation feedback.
Small caps won't make a big difference if the noise is coming from a strong source, which could easily be the case in a CRT. The only solution to that would be more shielding and twisting of wires and such. Though adding a high-capacitance low-ESR and low-ESL capacitor like a 1-10µF ceramic or film cap is still a decent idea to drop the high frequency impedance to ground. Doubt a 10-100nF would be good enough for an entire power rail, but putting them at strategic locations on ICs would still be a good start if there's some unwanted symptom you're targeting.

>>2493144
post schematics (if you have them) and pics or it didn't habben.

>>2493139
Good idea. Sure enough I got 15VAC across the rails so it’s very likely a bad diode. The working diode on the other side is a SKE 1/02 which comes up at a rectifier diode, I guess my best bet would be to cross ref that to a modern equivalent

>>2493151
>>2493150
>pic
I was shitting bricks while doing this.

>schematics
https://elektrotanya.com/sony_gdm-f520_chassis_21cr1_sm.pdf/download.html
Page 26. I'll post a picture of the area of interest in a moment.

Little bit more detailed description of the layout and problem:
This monitor is built modularly, where video is processed (=amplified) on a dedicated board.
This video board connects to the motherboard for power and some serial, but these connectors have been an issue. They flex and scrach, and these monitors even new would start to have rather serious overshoot/ringing issues caused by what i assume is resistance in the connector. And I've pinpointed it to be specifically the 12v rail which indeed goes to the preamps and main amps.

And yes, I've already tried various forms of deoxit, and it IS effective, but only for a day or two.

>>2493147
Okay at least I have the correct idea. I'm literally learning this shit on the fly here. But I'm wondering what size of cap I'm even supposed to use here, that ripple is like >100mhz.

>>2493167
This is the area I probed. You can see it already has shitton of random filtering components, and so do all the ICs. But it's clearly not enough.

>>2493168
and the capacitance of that capacitor is 0.22µF

>>2493167
Since you already suspect the board connectors you should fix them first, and crimp a new wiring harness or solder the wires directly to the board(s). Then check the solder joints around the flyback transformer and other connectors, reflowing anything that looks iffy. If any areas on the power board are darkened then you should check the resistors in the area and any electrolytic caps that might have been overcooked.

>>2493168
Is there a difference in the 12VDC waveform after the connector and on the analogue circuitry, compared to right out of the PSU itself? As the other anon says I’d want to replace the connector entirely. Also ferrite beads.

>>2493178
>>2493192
>Since you already suspect the board connectors you should fix them first
I've tried to do so in many ways. I've used all sorts of contact enhancers, I've tinned the contacts, etc. I've looked into replacing the connectors, but nothing suitable is available.

As I said, these units are known for having this issue even brand-new. And this specific model is also rather valuable when they work right. It is the best CRT monitor Sony made.
So I really want to create some reasonably easy patch for it, so other people can fix theirs too. If just slapping an extra capacitor on it can fix it, why not?

Sony also used this exact chassis design in many other monitors, with the exact same connectors and all without issue. But those had slightly different video boards, and never had this issue.
So even if I did find new factory connecotrs for it, it's likely they'd have the exact same issue as these originals. The filtering on the video board is simply insufficient.

>>2493192
>Is there a difference in the 12VDC waveform after the connector and on the analogue circuitry, compared to right out of the PSU itself?
I've not probed it, but I think we can assume it is. Either way, there's enough distance between the two that filtering is necessary, and that's what Sony thought too since there's such a mix of SMD ceramics, inductors and ferrites on the video board.

>>2493165
>cross ref that to a modern equivalent

probably you dont want an exact replacement for a dead diode.
coz the conditions that killed it could kill the replacement.
so, you should choose something more sturdy.
i like the 1N5404, good for 3A up to 280V, so it's useful in all kinds of scenarios.

>>2493198
>I've tried
Have you tried pushing on the connectors to get noise on your scope? Did you clean the adjustment pots and check their solder joints?

>>2493210
>Have you tried pushing on the connectors to get noise on your scope?
Not sure what you mean?
>Did you clean the adjustment pots and check their solder joints?
This monitor doesn't really have pots, it's all digitally calibrated. There's microprocessors in every corner. Even the convergence is tuned by dragging points in grid in a program, not by taping magnets on the tube. It's exactly as expensive as it sounds.
As far as I can tell everything else in the monitor is working as intended. It's quite low-hours, and in good condition.

>>2493214
>Not sure what you mean?
Gently move the wiring harness or ribbon cable while watching your scope. Tap on suspect connectors too.

>>2493062
wow. lol. that's cool but way too complicated for the job. may as well pick a different opamp.

>>2493138
>Why do you need RRIO?
Output is not critical as long it can fully drive the mosfet. Input? Hmm, I think
it has to be able go to the ground since Vref could be as low as a few mV.

But I picked this opamp primarily for its low Voff < 5uV and good phase margin so it doesn't need to be compensated. i also have a bunch of OP07s but those are producing some weird results I can't explain. There is a steady 500mV difference between V+ and V-. Can't explain this offset. Makes setting Vref harder.

>>2493215
Ah, my oscilloscope nor my probing setup are good enough for that. There was already a lot of noise and ironious samples, as my ground leads are bit long for ~300mhz stuff. Half the monitor is lacking its shielding so who knows what garbage it's picking up. Just moving around it affected it.

But I can still say for a fact there's no bad solder joints or intermittent connections, because the monitor works perfectly except for this amplifier ringing. I have tried pocking around with a stick seeing if the sympoms clear up, they don't, picture remains perfectly stable no matter what I do to it while it's running.

>>2493219
>affected it.
affected the oscilloscope, not the crt*

>>2493198
>I've looked into replacing the connectors, but nothing suitable is available.
Replace with DB9/15/etc. Good for cable-mount or chassis-mount or even board-mount. XLR is decent if you only need 2 or 3 pins. 8P8C/6P6C might be usable.

>>2493218
0.5V is a bit too low for it to be the clamping diodes (pic related), but the input offset current is in the realm of nanoamps. Sure you're not doing something stupid with the trim/offset adjust pins?

>>2493229
>trim/offset adjust pins?
No they are floating, but if I mess with them then I can make V- more or less track V+. Weird.

>>2493229
These are not trivial to replace.

I'll just try soldering different value caps on it tomorrow, this filtering stuff seems more like an art than a science. It already has so many random filtering components in there, there's no telling what the optimal values should be. Sony themselves seemed to just spam 0.22µF capacitors and 100µH inductors everywhere they could.
Good night!!

>>2493236
replace the y cap

>>2493236
Ah they're board-to-board connectors. Yeah those won't be trivial to replace, though if you desolder them (potentially destructively) you could replace them with ribbon cables.

>>2493235
this may be the answer, from ti.com:
>Can OP07 operate from single supply?
>Yes a single supply can be used. The output and input voltage range will not include ground.
>Input range will be 2V to VCC -2 V and output range will 4V to VCC -4V with a 2k load (to mid-rail)

So maybe I need a dual supply for this task.

why does my breaker keep tripping so far under it's current rating? I have 2 computers, a standalone ATX power supply and other accessories on a 20 amp circuit (router, monitors, etc) all pulling a grand total of 450 watts under typical/high load as measured by a killawatt. there is nothing else on the circuit, it goes only to the two bedrooms I have and the other one has literally nothing plugged in. there's a about 200 watts on one plug and 250 on another (whole different outlet, not from both sockets of one plug) and all of the sudden the breaker keeps tripping after just a couple hours of everything running. I'm not gaming or running anything intense, one computer is my desktop the other is a nas and the standalone PSU is to provide extra power for the ton of hard drives I have in the nas. no gaming with a 3090 or anything that would cause some huge transient or something, it's literally just a near constant 450 watt/3.75 amp load.
any idea wtf is up with this? getting real fuckin annoying having everything just shut off while I'm in the middle of downloading or compiling or whatever and having to start over. house is only about 4 years old so there shouldn't be any problems with the plugs/wiring/breakers.

>>2493281
Is everything plugged into a power strip/surge protector? Maybe it's a piece of doodoo.

>>2493283
i have a computer with monitors on one and everything else on the other (200/250 wtts). neither have ever given me problems in the many years ive used them. how would a bad power strip cause the breaker to pop other than an obvious quick short cirvcuit?

Guys I need help. I have two sconces on the front of my new home. neither of them were functional so I replaced them today. Now, if both are installed the lights flash. If only one is installed it works fine.

Why is this happening and how do I remedy it? Turning off/unplugging everything else on the switch doesn't help.

>>2493287
What sort of light are they, LED?

>>2493289
Yes LED and it comes as a "unit" I cannot just unscrew them and put a lower wattage light in.

>>2493291
And they hook up directly to mains AC?

>>2493287

if you have a dimmer or timer designed for incandescents, that can cause weirdness on LEDs.

>>2493294
Yes the standard white, black and ground wire wall sconce set up.

>>2493291
You probably wired them in series instead of parallel. Somehow. Maybe one is faulty. Or lose connections in your wiring

>>2493298
Get some Wago splice connectors and ditch the wire nuts.

>>2493199
>>2493165
>>2493122
The reason I said "fit through the holes" is because the 1N54 series and the 1n4003+ have wires that are considerably thicker and may not fit through the pcb holes.
The interesting thing is that diodes usually fail as a short but yours seems like it may have failed open. Check the filter cap too.

I've got this car voltage inverter which the transformers alongisde some mosfets seem to be shut. I have no idea for which transformers to look for. Are they outputting 110v? I'm not sure because the output is managed by a switching supply and the transformer is behind it just bumping up the voltage.

>>2493365
A diode failed short may have enough heat going through it from all the extra current to go open. Assuming nothing else breaks.

>>2493390
Judging by the size of the transformer, it will be switching high-frequency into that transformer with one bank of transistors (probably 2), then rectifying that to 160VDC or so, then switching that to AC with a second bank of transistors (probably 4). So you should be able to probe the rectified output of the transformer to find that HVDC. If you do see it, then the 60Hz inverter stage is buggered. If you don’t, then the high-frequency stage is buggered. This stage, as well as the 60Hz stage if it’s a pure sine inverter, will probably want an oscilloscope for live troubleshooting. That said, continuity measurements through the transistors may tell you which ones are good and which ones aren’t. You may get erratic results since you’ll be measuring in-circuit, but for power FETs I think it should be fine enough. Just wait a bit with your probes on the part for any capacitors to charge up.

Probably repairable either way, though you can get inverters for $100 so it may not be worth your time.

>>2493138
>Post circuit
it's the microphone i have posted here before, i just can't get that fucker to work.
At one point i had it working pretty ok, just with a bunch of noise and now i can't get it to output anything, only if i blow air at the mic capsule i can hear the sound output.
i'm starting to think the op amp or the tranny is just shit since i bought it from bug men as they aren't manufactured anymore
The usb ADC works fine, i tried connecting it to my phone via 3.5mm jack and it produces nice and clean sound, so that is not the issue.
Honestly it's fucking frustrating because i am not building a particle collider it's one fucking op amp that is supposed to boost signal from one fucking tranny, thats it.

>>2493415
>i'm starting to think the op amp or the tranny is just shit since i bought it from bug men as they aren't manufactured anymore
May be the case, but if it's mains frequency noise I'm betting it's an external issue instead. Even if the IC's PSRR is shitty, that would give you a hissing noise from the rail inverter, not a mains hum. Have you tried different grounding situations? Like plugging it into a laptop that's running of battery? Or connecting its chassis directly to a copper tap or ground lug?

Is the metal chassis connected to the circuit's 0V?

Also that's an instrumentation amplifier, not an operational amplifier.

>>2493404
Well the transformers I think are busted, there's no continuity between input output at all. Also there's at least 1 busted mosfet on each input bank. I have nothing to measure. How do I look for transformer replacements?

>>2493420
There should be a finite resistance through windings, in the red zig-zags. The HV side and LV side of the transformer should always be fully isolated from one another.

Also those bottom 3 pins look like their solder is melted together, if you get less than 5Ω on the DMM I'd look into that. There should be no solder bridging in the yellow lines. The blue hatched area looks like it may be a broken trace.

>I have nothing to measure
What do you mean? Plug it in. There should be a high voltage between the cyan and magenta marks I've drawn.

Is there a fuse?

>https://www.farnell.com/datasheets/2064131.pdf
>single tristate pwm input
uuh, how do i split that into two mcu pins? connect the two pins with 1k resistors? ill have to figure out the 3-phase pwm waveforms.

>>2493419
>but if it's mains frequency noise
thats to the main issue now, basically for some reason i'm not getting any sound out of it unless i blow at the mic, where then i can hear very waint blowing sound on the output.
And yet i checked all the wires and shit and nothing has changes so not sure what the problem is.
As for shielding, back when it sorta still worked i tried wrapping the capsule and the power source in grounded tinfoil which had no effect on the 50hz noise.
Powering it from isolated lab bench power supply reduced the noise.

>Is the metal chassis connected to the circuit's 0V?
you mean the mic cage? yes

>>2493428
No audio at all now? Yeah that's probably the shitty IC. Buy them from actual vendors, or make your own instrumentation amplifier from quality op-amps if you can't get monolithic inst-amps locally.

>>2493429
>No audio at all now?
No normal level of audio only blowing on the mic (to max out its volume) can be faintly heard.
Can't buy from normal vendors because they either don't sell it or they cost like 10 times more than chink ones.
Can't just replace it with different model either because i don't know enough about amps to pick one that works well with this capsule and tranny

>>2493431
Any old high-quality or audio-quality op-amp should work fine if you put 3 in an instrumentation amplifier circuit. I can get OPA2134s locally, they should do the trick.
I can also get INA126s locally, which would work without needing 3 of them. Good JFETs are harder to come by, but I’d try any you can find.

>>2493431
The whole circuit is needlessly complicated. There is no need for split supplies.
https://www.ti.com/lit/ug/tidu765/tidu765.pdf
You need to make sure that jfet is biased correctly, there must be resistor from gate to ground otherwise it will just be floating. Here is a better guideline on how preamps for condensers are designed
https://www.youtube.com/watch?v=niZizzHBanA
For electrets same things apply, just that voltages will be lower.

>>2493438
>no need for split supplies
There is, a wide voltage range is important for JFET circuits and getting properly linear operation out of the amplifier. Having a central ground rail is also beneficial for the jump from differential to single-ended signal, though it’s not essential. Rail inverters are standard enough that I’d likely make a similar decision. You could design a circuit to use just the 5V supply of you found a good enough amplifier IC, but you wouldn’t be able to use JFETs or JFET-input op-amps, so it may not be good at all.
>there must be resistor from gate to ground otherwise it will just be floating
In this case, the electret microphone capsule provides a natural negative bias relative to its grounded other side. It’s a bit strange, but it worked well enough in the original video. Tying it to V- with a high value resistor might be beneficial (after an extra power-supply RC filter) but then again the thermal noise might make it worse.

>>2493434
>>2493444
So there is this one suspicious behavior i get. During operation the ina217 gets hot enough where i can't keep my finger on it (even back when i could still hear my voice with noise)
The whole pcb draws about 160ma.
When i remove the ina217 it's about 60ma.
So the ina217 is using 100ma then? that seems like a lot. I am pretty sure it is not normal for it to get this got during normal operation.
I tried connecting 15V to just the bare chip without it being plugged into anything, and it only draws 20ma.
Could the USB analog to digital audio interface really be drawing so much current from the amp?
When i tried hooking the ADC to my phone audio 3.5mm output, it worked no problem

Ok so i made a 7-segment display driver using a MAX7219 and it works great with red 1" common cathode 7-segment displays. I bought a few others which are green, blue, orange and white.

Everything except the blue and white works. Any idea what could be causing the fuck up ? The IC is getting 5Vs, 2A from the power supply, Iset register is set to its max (around 40mA). Are they drawing more current than what the MAX can hold ?

>>2493424
Yeah the input is fused with 3 30 amp fuses and they're all blown. Thanks for noticing the bridged mosfets, I didn't and for clarifying the transformer, I knew there was an impedance between input and output, whatever. As for the resistance through the windings there's 0 ohms on the output and the diodes are all ok.

>>2493447
hmmm tried hooking up oscope to the J3 microphone connector on the pcb, to see what was coming from the capsule and i get pretty much a flat line, unless i blow on it, then i get verly slight wobbly line like this.
But i think thats maybe normal since the electric signal coming from the tranny would be super weak right?, hence why i need the opamp

Tube noob here, tried out various caps on it. 100nF, 2.2nF and 0.1nF. None quite fix it perfectly, but after I let the monitor warm up with the 0.1nF cap, the picture ringing almost entierly cleared up. But a trace of it still left. But my idea seems to actually fucking work.
I think my cheap aliexpress ceramic disks don't quite have low enough ESR to filter out 100-300mhz ringing effectively, I think I'll order a set of proper SMD ceramics and see if I can completely clamp the problem down with those.

But I'm rather surprised I actually managed to do this without killing myself or any of my hardware.

I was worried that thin copper wire (0.1mm) could introduce a significant resistance but even at 10cm a website says it is 0.2ohm. would it be really that low?

>>2493532
You could just measure it yourself. 0.2 ohm isn't hard to measure. Here's a video showing how to do that without using a miliohm meter:
https://youtu.be/0vMvCVyOp9g

>>2493543
I don't actually have the wire yet, wanted to ensure it will work for my project. My resistances needs to be somewhat accurate

>>2489838
Found a broken solder joint in this motion actives halloween decoration. What is this thing, and how do I fix it? I’m guessing I want to solder it to the edge, avoiding the center disc.

>>2493550
Other side

>>2493550
>>2493551
it's a piezo disc. it makes noise.
beep boop

>>2493553
>it makes noise.
or detects vibration

>>2493553
>>2493554
Yeah, I guess this should been in stupid questions. It looks like edge-soldering is the way to go, correct?

>>2493558
Yes.

>>2493558
Tin the wire and add a tiny amount of flux so you can minimize heating time on the disc.

How do I solder aluminum wires? This shitty Chinese USB camera stopped working half of all times. When I replaced the USB plug (awful A/Mini-A hybrid connector), it started working reliably again. The new plug uses copper wires. (I just cut off good cable and connected the wires.) I can't solder the aluminum to the copper, so the wires are just physically intertwined.

>>2493566
PS: the shitty Chinese device uses aluminum wires in the USB cable, but it's impossible to open the device and replace the cable completely. Fucking trash, cheaping out on the USB cable and selling tech outdated by 20 years for 10x the price.

>>2493566
Try a higher power soldering iron or bigger tip, wood burner (lmao), or torch/torch lighter. Maybe a splice connector. Then you have to shield the whole cable again.

>>2492428
step by step hand holding of pcb diagnosis, you're not even measuring components going backwards from the output side towards signal in.
this is going far beyond asking for help, you want others to do your job for you

>>2493569
>torch
Even if I had any of this, I'd char the whole cable. I don't have anything to splice/crimp it either. Maybe I'll drown it in hot glue to avoid that the twist-connection gets lose.

>>2493568
>but it's impossible to open the device
Get yourself an assortment of guitar picks (light, medium, heavy, etc) and use them to pry and shim.

>>2493564
Halloween is saved

>>2493576
It's one of these. Literally a pipe, cable goes in on one end, sensor is on the other end. The thing on the cable can be rotated.

>>2493586
Noice. Good job, Anon. Maybe wrap the exposed part of the wire with a short piece of tape just in case.
>>2493587
Okay, I see. You'd have to wreck it to open it. Pretty annoying. Check out pic related USB breakouts.

>>2493597

>>2493548
What's your project?

>>2493570
Thanks I'll be sure not to ask you for help next time I post in /ohm/

>>2493280
Sure enough, dual supply works like a charm.
But now I am thinking what if I just set the external Voff trim to something extreme, such as -2V? Wouldn't that have the same effect of nulling the V- V+ difference? Except I haven't seen this mentioned anywhere in the application notes or in discussions so maybe that's not going to work.

>>2493447
A precision amp getting hot is not intended. Replace it with another one (you did buy multiple, right?) and disconnect the output. Just to ensure there's no high-power load. If it still heats up, you got chink'd.

>>2493448
Dropout. You're running on 5V, the white and blue LEDs are taking 3-3.5V, which the IC probably isn't designed for. Old designs like that may well be using TTL logic, so you can expect the BJT dropout on the anode and cathode to add up to something significant, especially if they're using darlingtons.
Also there's a graph in the datasheet relating output current to output voltage.

>>2493451
>there's 0 ohms on the output and the diodes are all ok.
That may well mean a dead transformer. You may want to make a milliohm meter (just a battery + LM317 + resistor + your DMM) to trace where short circuits are actually happening.

>>2493454
Normal behaviour, that's why you need the amplifier.

>>2493566
Olive oil + sandpaper to get rid of the oxide layer is what people do on solid aluminium, but for stranded you'd be pushing your luck. Look up if zinc chloride does it, if not then use crimps.

>>2493575
twist the wires together, then wrap it with some solid-core copper wire and squeeze hard down with your pliers to set it in place. Or just buy butt terminals or whatever.

>>2493628
>Wouldn't that have the same effect of nulling the V- V+ difference?
Yes it does. So long as you respect the total voltage difference and the common-mode input range you'll be fine.

>>2493466
>>2493144
I even probed it to confirm I'm not hallucianting. Halved the ripple.
Changed to a 150pF one and it seemed to improve even further - after warmup. It always takes a few minutes to clear up. But that's actual stock behavior of many F520's, from what I've heard.

>>2493655
>(you did buy multiple, right?)
nope, one costs 5 bucks so i can't afford to buy more than i need
but like i said, when i pulled the amp chip out and hooked it to power by itself it only drawed 15ma, which is about what the spec sheet says, so the amp itself is not busted

>>2493601
It is a toy project, I want to make a very small qwerty keyboard with full qwerty of buttons. Pic related, imagine TP is ADC input pin for my MC. So I will have groups of such buttons and I will check the voltage levels of their output to see which one in that group is pressed.

It looks like I have a reasonable accuracy on my ADC with 10 levels between 0v & vcc. So I will group 10 buttons like pic related and use 6 of them for a full-ish qwerty keyboard.

I was worried that the thin copper wires would add significant resistance to fuck up my calculations and thresholds

Eventually I want to build a very small "desktop PC" using such small components I made. It will be consist of a 1 inch "screen", a computer case where my main MC will be in and the keyboard I am making. Then I will write some software for that PC (I am mainly a programmer so that part will be the most fun)

A mouse would be nice but it looks like it is would be a lot trickier

Received my LM386Ts.
How can I test if they are genuine?
I've seen a lot of 386 offers had bad reviews claiming they were bad. They definitely don't look new. Seems like the old printing got overpainted or something.

But the pins don't short out. Can I find some reference as to what resistance they are supposed to have?

>>2493669

>>2493655
>Yes it does. So long as you respect the total voltage difference and the common-mode input range you'll be fine.
my only concern is, would it require some tedious calibration? the default Voff is a few uV with the external offset pins floating. Now I am injecting some random voltage that I can't really measure accurately. Wouldn't that kill any advantage of a precision opamp?

>>2493444
Source follower will have more than enough swing with acceptable linearity, because input signal is so small ~100 mVpp at very loud inputs (100dB). If you're just recording speech it will be even smaller so worrying about split rail biasing is kinda pointless. If you want better linearity you will need to use feedback as shown in video I linked. Just increasing supply voltage won't help you get rid of intrinsic non-linearity - you might get larger swings, but they'll still follow square characteristic.
>In this case, the electret microphone capsule provides a natural negative bias relative to its grounded other side.
As far as I've seen some people add resistor to ground and some don't and rely on leakage paths. The question is how does this affect stray EM noise pickup? Adding parallel resistor won't add much noise because it will get filtered by capacitance of the microphone. And since usually resistors are in 1G range cutoff somewhere around 5-10Hz.

>>2493669
Why not just buy from a refutable seller?

>>2493710
>refutable

>>2493713
Reputable. Maybe digikey.

>>2493716
digikey
mouser
newark/mcm
marlin p. jones
jameco

>>2493720
LCSC

>>2493710
>>2493716
>>2493720
What is the deal with aliexpress? How can they sell things cheaper than such refutable sellers? Do they buy directly from factories/producers?

Or if it is cheaper then it is fake and possibly does not adhere to its original's specifications

>>2493724
They salvage parts from dead devices, relabel low quality parts as higher, generally scam any way possible.
Look at >>2493673 and tell me that is in any way new or looks like it came from a reputable source.

>>2493727
Yes we reputabre source! Good prices, buy our dog chip! I give discount! 2 for price of 1. What want more? Radio chip, hi-fi audiophire $1 only, quarity guaranteed all new top from factory. Give 5 stars now! We happy you happy.

>>2493673
Those are fakes. You could've just bought the lm3886 pcb kit from china and got the same fakes included with it. They do work though, I've tried them. Not sure what chip they really are.
I stocked up on lm3886, lm1875, and lm4780 before the chip shortages.

I have several gyrators going on in this circuit and when I don't feed it with a real dual power supply (I was feeding it with a single source plus resistor divider midpoint), the gyrators seem to affect even the very input before the buffer, producing kinks in the bode plot that shouldn't be there. Is that something that is supposed to happen?
Vb, by the way, is just ground at 4.5 V

>>2493733
Enjoy your chinese fakes.

>>2493745
If you have proper biasing there shouldn't be any difference between having split rails or just shifting everything up by half of the supply. If you're using resistors to get midpoint you might be loading divider too much which would result in weird behavior. Does the same thing happen with a plain voltage source instead of divider?

>>2493727
>>2493737
>fakes, salvages that work perfectly fine
What's the problem then? I have rare Japan made power transistors that I bought on aliexpress dirt cheap. They work fine. 100W no problem. less than a dollar a piece. Who cares where they got them as long as they are genuine parts that work. Same with IRF250 and other high power MOSFETS/BJTs.

Anyone using lithium iron phosphate? 3.5V charging voltage vs 4.2V for li-ion and lower energy density, but immensely more durable... according to Wikipedia 400-1200 cycles for Li-ion, 2750-12,000 for Li-IP. Tolerant of both overcharge and deep discharge. The way Li-ion just dies just sucks

>>2493763
Not OP but "perfectly fine" is hard to prove. It may look like it works perfectly fine for your usage but possibly they don't satisfy all the specifications and would fail for someone else that pushes it a bit harder. Or its life expectancy is lower than it claims

>>2493753
The problem disappears when I use two voltage sources in series with the middle node tied to ground.

>>2493677
Shouldn't do, but I've never played around with op-amps that have those extra pins. Leaving them floating shouldn't be an issue as they're low-impedance.

>>2493701
>Adding parallel resistor won't add much noise because it will get filtered by capacitance of the microphone
Would it really filter 60Hz that much?

>>2493745
>2kΩ + diodes to GND from op-amp
Yeah you're going to want to buffer that GND rail.

>>2493768
I'd probably use them more often if there was a cheap TP4056-tier chip that charged them for me. Thermistor required, don't charge them below 0C.

>>2493777
This. They may well be QC rejects.

>>2493777
That's fine by me. Derating is common practice anyway, so I just test at full or close to full power once to make sure it is not a total dud since there are horror stories about "fake tiny die that can't take much current". If everything looks OK, I use them at 30% of their power. Say 30W instead of 100W. Derating them heavily is no problem, since they are cheap and I can parallel a few of them. And I am laughing at autists checking the markings attempting to verify the etchings with a microscope or testing them with a magnet or cracking them open to measure their die and other stupid shit. Spotting fake transistors seem like a sport.

>>2493710
Because they are fuck expensive? This particular seller didn't have any negative review, well, only "arrived, did not check" tier retardation that you know from ChinkExpress.
Anyways, I still didn't get my question answered. I want to check that it works at all before I build a PCB or something.

>>2494009
What kind of parts are you buying? Are there no cheaper but electrically equivalent parts, LCSC or otherwise?

>>2494009
They're probably salvaged, but how the fuck are we supposed to know if it works? Test it before soldering.

Can somebody explain what do the Ts and Ps mean?

>>2494047
Oh wait, the T means turns right? Then what is P

>>2493425
Ok so I've thought of these three methods to get tristate PWM from 2 pins. Method A gives 0.5*Vcc as an output when A is high and B is low, which may not be accepted (idk just says voltage thresholds, not impedance thresholds). Method B will give high-Z when A is low and B is high, but I'd need to throw both A and B up and down, which may get annoying for bit-masking and such. Method C gives high-Z when A is high and B is low, I think. The switching speed of the FET won't be as fast as I might want, but I doubt it makes a difference as turning off all FETs during normal operation results in current going through half-bridge body diodes anyhow. Which is odd considering a non-PWM waveform usually has that modified-sine dead-zone in the middle.

>>2494041
By comparing the resistance values of the pins?
Is there some chart or something?
The LM3886 is a quite popular linear audio amplifier.

>>2493737
There aren't many 3886 pin compatibles, are there?
If it's just a salvaged chip I don't see a problem. And I doubt Chang can just brew together semiconductors in his basement.

>>2494063
>By comparing the resistance values of the pins?
By plugging it into a breadboard or soldering some wires to it and getting it to amplify something.

>The LM3886 is a quite popular linear audio amplifier.
And because it's popular it's been hit by the chip shortage. I always go for obscure parts that aren't on arduino boards or aliexpress modules exactly for that reason.

>>2494066
Coulda been run really hot, but generally I'd agree. I'd just derate it as the other anons said.

>>2494066
Here's a picture of two different LM3886T chips I ended up with. I'm inclined to be skeptical as to the authenticity of both of them.

>>2493727
>They salvage parts from dead devices
then why is there never solder on their leads?

>>2493447
dumb question, but did you get the pinout correct? I see in your pic it goes "2-1-8-3 etc.", and if you assumed that was the physical pinout you probably wired something backwards

>>2494180
>why is there never solder on their leads?

every chip you buy, new or used, has ''tinned'' leads, i.e. solder.
and chinks have developed procedures to restore pins, and redo markings to make them look new.
this is an industry, not some farmers who solder-suck in the evening for extra cash.

>>2494180
>add flux
>melt
>bang on table
>clean
wa la no more solder

>>2494185
it's def correct.
at first it worked, i could hear my voice nice and loud, just with a lot of background noise, but now it is picking up only if i blow on it to max out the volume.. the only choice i have now is to try to make a new pcb and resolder everything to see if there was some bad connection or something somewhere
other likely candidate would be the opamp, since it's chinkshit it's possible i got chinked and.. wait a minute..
i just realized something right now, when i change the gain resistor it has no effect whatsoever on the circuit.. which really puzzled me how can that be... but.. what if that is happening because the IC isn't INA217 at all, but the ultimate chinkage where they take cheap ic and print INA217, so i in fact got something like a fucking lm324 opamp, that they printed INA217 on and sold it to me for five bucks.. holy shit this makes so much sense
FUCKING CHINKS!!!

>>2494282
fucking hell
i got chinked after all
https://www.aliexpress.com/item/33020271144.html

god i am so fucking wasted i spent SO MUCH FUCKING TIME on debugging this shit
now i have no fucking idea the 2N4416A i order from the same seller is legit at all, or if it's also some useless piece of shit tranny like 2n2222 with2N4416A printed on it
fucking bugmen deserve to get nuked

>>2494282
>>2494283
>>2494284
you pay, you go! i feer sorry for your mudda.

>>2494154
Mine looks more like the first one but the label looks completely new. Like they overpainted it.

>>2494285
Lel that made me chuckle

>>2494284
Sometimes I'm glad to be a retard living in a western European socialist country.
I do my trade certificate in some retard job workshop and I get all the high quality parts for free.

>>2494290
Couldn't find an LM3886 though so I had to order that one. But basically I have a ton of free time in that workshop and thought I'd build an "audiophile" amplifier.
Tell me what else to build, Anons.

>>2494289
We'll have the last laugh, Anon. Chiner in crisis. Tianjin times ten. lmao

I can NOT find an internal transistor schematic diagram of a CD4053. I just want to compare it to the CD4066. What’s the difference between an analogue switch and a bilateral switch anyhow?

>>2494318
>difference between an analogue switch and a bilateral switch

two name but same one ting.
like sneaker, gym shoe, trainer, tennis shoe, runner, kicks, nig wedding shoes, dunks, chucks, etc

>>2494284
buy it cheap, buy it twice

I got a diy tiny tesla coil and fucked it up by accidentally swapping mosfets with each other. Then probably further fucked it up by trying to desolder them and heating too much

How can I make a bluetooth jammer? I hate my neighbor

Retard here
I have an RF transmitter and I want it to feed signals into this antenna: https://linxtechnologies.com/wp/wp-content/uploads/ant-433-usp410-ds.pdf see page 7
How do I design this matching network if I don't know the input impedance, which would be the output impedance of the RF transmitter?

I know it says it's no requirement but I still want to know how this would work

>>2494355
Buy more transistors, try again.

>>2494366
Ask the ham general if you want to be told off for being an irresponsible retard. Do your own research. It’s possible that cheap Bluetooth transmitter ICs have enough control available to adversarially target existing channels, but you’d have to read some data sheets to figure out whether frequeny allocation can be completely manual. Bluetooth engages in channel hopping anyhow, so you’d either need to actively hunt for Bluetooth transmissions to target, or smother the whole range in noise. The latter gets you V& faster.

>>2494370
Take it apart and see what the final filtration components are doing. If it’s not a microwave hell of PCB waveguides it should be doable to just plot the components and whatever IC they’re connected to. If it uses discrete transistors for an output though, not sure how to gauge it.

>>2494366
BD_ADDR
spam_DoS_niggertits.bat

>>2494370
https://en.wikipedia.org/wiki/Antenna_tuner

>>2494047
>>2494048
I would assume P is strands of wire. As in 5 strands of 0.70mm diameter each.
Why P? And why is the P only on on side
of the transformer? And why do they list diameter rather than cross section? Because China, I guess.

>>2494386
has anyone clicked this link?
is this a link?
>it has a compelling cartoon frog image
>part of it is "spam niggertits.bat "
if this was meant to lure me into copy/paste a link then its working but still too cautious
it could be information and ideas I would use
>I hate the antichrist

>>2494518
.bat is a filename, not a url

>>2494519
should I use it?
will it grant me inspiration to disable some antichrist supporting electronic interference?
what is the of it?

>>2493109
Did you get the diode replaced yet? If not did you at least take that one out and test it with a multimeter?

will this power this? it is a 12v 10ah battery and the immersion heater is 120w 12v

business idea: a zero noise power supply: a lipo battery, that is being recharged from a generic power supply powered by mains in the background while providing power to noise sensitive circuitry. or two lipo batteries. the main and the backup, and switch between them automatically. any caveats?

>>2494679
It'll draw 12A, in a 5 parallel cell configuration that will be 2.4A per cell. Most half decent Li Ion cells should easily handle that current, but the specs clearly read 10A, which is the max rating for continuous use, so the battery protection will probably kick in after some time.

>>2494679
From the low res image I can't identity the capacity. It doesn't say 10Ah, just 10A, which I assume is the max continuous output.
At least provide better images or a link next time.

>>2494696
>>2494697
guess it works but for how long?

>>2489838
Hello anons, i don't want to make aa thread for this stupid ass question, so bare with my stupidity:
i want to make an aux cable that has volume controls and next/prev song buttons, BUT i want the cable to have a knob instead of buttons; now, i kno i can do one for volume just fine, but is there a way to add the buttons of control / add gestures to the knob for controlling stuff? reason i'm doing this is because i'm retrofitting a cheap radio to my f-100 and i want to connect my tablet to use the music on it; but the cheap shit radio doesn't have a volume knob, it's all in the screen and it's awkward so i was wondering if there's a solution? i'm already using the cable of some random headphones and it can control everything perfectly but the buttons are minuscule and 'i dlike a big ole knob; video related:
https://www.youtube.com/watch?v=decGiV7eFSI&ab_channel=HardwareUnknown
This dude made the button i want to achieve at but his can only control volume

>>2494731
Steal a car battery.
>>2494734
Solder your tactile buttons and a dual-gang pot to a piece of protoboard, then solder the in/out signal wires. Put it in a small enclosure like an old flash drive casing. Done.

>>2494734
nevermind I misread your post
You're describing a digital rotary encoder with a tact switch under the knob. You'll need a microprocessor to translate the encoder output to button presses on the radio.

>>2494736
i had camp cooking in mind but cooking off of wood is probably more energy efficient. i dont think a 50w solar panel will do much in north western europe

Correct me if I'm wrong, but shouldn't a 3.3V square wave (Vrms = Vppk/2) with an output impedance of 50ohm be able to deliver 33mA worth of current?

The Si5351a digital frequency generator claims it's 50ohm at the output, but also claims it delivers a maximum of 8mA. Doesn't that mean it's closer to 200ohm output impedance?

>>2494738
thank you, i'll look into that, ia lready made the volume knob with a rotary encoder, but i'm struggling to add the (next, prev and pause) buttons, do you know if i can just solder additinal buttons to the thing?

>>2494739
That's 50W in full sun, and still not enough to do much other than charge phones/tablets/laptops/small battery packs and run a few lights (weather permitting). Heating elements will drain your battery too quickly, so save the electricals for small stuff and cook the old fashioned way.

>>2494746
>do you know if i can just solder additinal buttons to the thing?
So long as the microprocessor has enough inputs.

>>2494746
https://www.amazon.com/android-media-remote-control/s?k=android+media+remote+control
Maybe these will work for controlling the tablet. I watched the video you posted and saw that he's using a plain analog pot (easy), not a digital encoder (more involved). You can use either one, but analog is a peach of cake.

I want to make a simple CB (27MHz) morse code transmitter, doesn't need to be powerful, just within my room.
What would be easy way to make some simple tunable (vary thru CB channels) approach ?
I find solutions like colpitts oscillator a bit wild (hard to get practically accurate).

Is there different way to get precise frequency control besides going FPGA ?

>>2494803
hack a cheap RC car transmitter

>>2494805
nah, want diy my own board.

>>2494734
If the tablet has a TRRS plug that can support an inline microphone, then it can probably also handle inline controls. Both for volume and for pause/play, maybe previous/next also. See:
https://electronics.stackexchange.com/questions/242883/how-do-media-buttons-on-a-plain-wired-headset-work
Should be pretty easy to replicate this, just ensure you keep your cables well-shielded.

>>2494803
I think the si5351 goes that high, otherwise a phase-locked-loop with a VCO of your choice should give really accurate frequency setting. There should be DDS ICs that can go that high without difficulty. Look at a cheap CB radio schematic to see what it does for a clock source.

Then use a Class-C or RF Class-D amp to turn that low-power square into a higher-power sine. Antenna selection is magic that should be divined over at the ham radio general.

>>2489838
does onsemi's online store work? i added a free sample to my cart and click "check out" but it doesn't fucking do anything. tried different browsers, different PCs, etc.

>100x 2n2222a
>$5
What do you think, ese, too good to be true?
https://articulo.mercadolibre.com.mx/MLM-949979348-100-pieza-2n2222a-transistor-npn-_JM

>>2494864
Par for the course
https://lcsc.com/product-detail/Bipolar-Transistors-BJT_Foshan-Blue-Rocket-Elec-2N2222A_C358533.html

How can I shift the center of gravity of something electronically? In 2d? I want to tilt something front or back or left and right. Is there a component name or something that I can google?

How does a battery charger test battery voltage to switch between CC and CV etc, especially if battery is quite drained. It could measure a higher than actual voltage if idle, but once you connect even a tiny load, the voltage will sag, probably because the resistance of discharged battery is very high. Does the charger connect a load across the battery and then checks the voltage?

>>2494948
Don't do it, Anon.
Don't put a bucket of poo above the door frame on a remote controlled tilting platform.

Do these cheap phone chargers only provide constant voltage and the actual charging profile is maintained by the phone circuitry?

>>2494971
It is the phone or more like a circuit on its battery.

Whats the best way to add an LED to a monolithic voltage regulator to indicate when load is connected and it is supplying current? Shunt resistor + High side current sensor?

>>2494948
Linear motors (or motors with a lead-screw) would be able to move a mass from side to side. Rotary motors, likely ones with a gearbox, would be able to move a mass around some central point. Both could be made to work, but if you want stability then the feedback loop would need to take into account the moving inertia of both the whole thing and the moving weight, and the force coupling the two. A cheap RC servo may be the easiest way of doing this.
More detail if you want a more concrete answer.

I simplified the circuit of the DOD FX-86 distortion effect, which is basically a modified Boss Metal Zone.
All unity gain buffers were removed. All bjt-based gyrators were replaced with op amp gyrators. A new feature is a buffered ground at half supply.

>>2494983
Or perhaps just a big ass 1A - 2A LED and adjust for the voltage drop across it? As long the max current consumption is known, that would probably work fine?

>>2494985
Neat. Put a cap on R31, and ensure all input signals are relative to the buffered signal. A downside of not using he negative rail as ground is that you can’t use the built-in switch in an audio-jack to turn the thing on, but you may still be able to use a barrel jack’s switch to enable battery use.

>>2494990
Won’t have a good threshold. An easier way would be to put a series resistor before the regulator and have a base-emitter junction in parallel with that, which is capable of passing the full current. You then use the BJT to turn on a lower-power LED circuit (large value resistor). If you set the resistance to be 1Ω, the transistor will remain off until 0.7A. Above that, the series current will prefer to go through the BE junction instead, turning the LED hard on. Should be within 20% accuracy.

>>2494971
If it's USB, then constant 5V, charge circuitry on the phone itself. Basically all phones, even the old ones with the cylindrical jack, will be this way. Though cordless landline phones that sit in a dock may have the charge circuitry inside the dock, not the phone, I'm not sure.

>>2494975
The circuit on batteries is a protection chip, but there will be a BMS on the circuit board of the phone itself. The BMS handles charging for the most part, the protection circuit is just the last line of defence for safety.

>>2494990
>>2494994
Pic related. Assuming you have the voltage drop for it.

>>2494999
>bean.png
Damn, thanks, so simple, I totally forgot about the ubiquitous Rs between b and e, it is used in so many applications.

>>2494999
>BMS
What about all these new USB Power Delivery super fast charging high power adaptors? Do they have any additional smart circuity or they are just beefy power supplies with higher voltage and current ratings? I read that USB-PD can supply different V/I such as 5v 3A up to 20V 3A or something like that. That's insane. Feeding 20V into the phone?

>>2495047
Found a video:
https://www.youtube.com/watch?v=DQLrZA5RMjQ
fucking hell

>>2495047
Still constant-voltage, but they start off at 5V then negotiate what that voltage is.

20V ones are usually reserved for laptop charging.

>>2495047
There are two parts:
1. the trigger board which is installed in the device to be powered
2. a USB-C PD charger
The trigger board can be hard set to a specific voltage, or it can negotiate with the PD charger for a higher voltage.

No seriously, how the hell do you push 5A at 20V through a little puny USB-C connector that probably have like 30AWG wires. This USB PD stuff is fascinating and the protocol is incredibly complicated. I'd rather have a hard switch than "negotiate" current and voltage. There are some cheap USB HD chips on aliexpress but. The last famous words before the explosion "Oh. I thought you wanted 100W". I wouldn't risk my phone or laptop.

>>2495252
>make a tiny software bug
>get voltage that's 3 times higher than your device can deal with
Also, as is standard with low level stuff, the average code quality is so incredibly shit that I'm genuinely scared of modern technology. I hope you hardware designers and electronics engineers are doing better...

is it possible to get 12v 10 amps from a battery pack, if so how?

>>2495258
*sorry i meant usb pd

>>2495258
I think it goes up to 5A. Getting 10A out of USB-C would be funny. Even 5A is pretty insane. The new spec claims over 240W. But they increase voltage not current to minimize voltage drop and heat. Up to 48V. Soon enough you can charge your Tesla over USB-C with the same cable you charge your Iphone with.

>>2495268
What's next, USB-AC with a dropper capacitor that gets rectified by the nigger board? lmao

>>2495268
yeah or you could use 12v car appliances

i'm trying to build my own electronically controlled load using a MCU to control everything. i understand that a DAC will command an op amp control loop, which then increases gate voltage of a MOSFET. IRFP250 is commonly used in commercial electronic loads, which has a 2 min, 4V max gate-source threshold voltage. most DACs are 3.3V with a more or less rail to rail output voltage. how can the DAC output be accurately and linearly stepped up to at least 4V, so as to reliably turn on the MOSFET?

>>2495272
You don't need 4V reference to drive the mosfet. The opamp will do "whatever it takes" and ramp up the Vout to drive the mosfet. In fact, if your sense resistor should be really tiny especially if you need high current, and the voltage drop on it would be on the order of mV. So your Vref should be really low if your Rsense is a low value.

>>2495272
>>2492988

>>2495278
ohhhh, i get it now. one commands the current (well, voltage drop across the sense resistor) which is likely <1V. that makes a lot more sense, thanks!

>>2494506
Is it bad if I put some tape around the core to hold it real good? I had no way of getting in there.

>>2494869
That site doesn't take oxxo pay though.

>>2495280
Yeah, think of the opamp as a black box that does whatever it takes to make V- equal to V+. Let's say you set V+ (which is your Vref) to 1V. Now the opamp will try to bring V- to 1V. How does it do that? It will drive the mosfet by increasing Vout until the voltage drop on your sense resistor is 1V so V- = 1V. So if you need 1A, you calculate it as Vref / Rsense and set both accordingly.

>>2495285
What I found out is that the high frequency part is made of 16 strands while the low frequency only of 2. Didn't unpack the whole low frequency to count the turns but there were a few. So T is turns and P is strands. I started unpacking it cause I measured 0 ohms on each winding but I realized it's alright for such a small size transformer.

>>2495257
>make a tiny software bug
Well there has to be enough of a bug that it sees multiple wrong bytes of identifier information. Not generally something that happens.

>>2495270
kek

>>2495280
In reality, the error amplifier will be comparing the difference between the DAC (maybe with a buffer after it) and the output of the current-sense differential amplifier. So you'd set the gain of the current-sense amplifier such that the maximum voltage you'll see on its output is the maximum DAC voltage. You may want to use 3.3V for this, but from what I hear DACs often prefer to work at a lower voltage than the supply, more accuracy or whatever. This would likely be 2.5-1.1V from a reference.

>>2495292
Who the hell is that? You wouldn't want to just buy those from LCSC as they'll charge you $40 in shipping. Only buy bulk orders from them. Do you have a local depot of RS or element14 or whatever?

>>2495414
The PD protocol works like this: the source sends a list of voltages it supports (with maximum current draw for each). The sink has to reply with the number of the list entry which it wants. Now add code monkeys to it, and marvel at how much can go wrong.

>>2495415
Is that even done in a microcontroller though? I figure it would be set in PMICs in a rather non-volatile way. Like OTP-PROM, or by pulling certain control pins high or low, or what have you.

>>2495417
Is there such a solution? I only know of the FUSB3xx chip, which leaves most of the PD protocol to the microcontroller.

>>2495421
>which leaves most of the PD protocol to the microcontroller
Oh, in that case we're doomed. Can only hope that problems get caught in QC testing.

thinking about building a tube amp dildo to look at when im high and with a girl give me something

What does +50-10% tolerance rating mean for a capacitor?
Does it mean it can have at most 50% more or 10% less capacitance?

Or 10 + 100 %? What does that mean?

>>2495437
>Does it mean

yes.

>Or 10 + 100 %?

the intern forgot the - sign in front of the 10.

How can I test a trimmer pot? This is supposed to be 100 ohms +-20% 0.1W

It is on the right output amplifier which runs hot - so trying to start diagnosing with a multimeter...

>>2495589
Desolder it and measure it. But if the amp isn’t self-destructing, find a power-resistor between the high-side and low-side transistors (there should be 1 or 2) and measure the voltage across it with zero input signal. Then calculate the “bias current” from that voltage and the resistance. Do the same for the good channel and compare. If the bad channel has a significantly higher current than the good channel, it’s a safe bet that the biasing circuitry is bad. Hopefully it’s using a proper Vbe-mult, in which case there are a variety of possible causes, but yeah the trimpot is a good place to start. Measuring the collector-to-emitter voltage of the Vbe-mult should also tell you if anything funky is going on.

Also you better try to find a service manual.

>>2495589
Why are the transistor tabs bent?

>>2495604
Thanks! I have the manual and I think they describe what you're saying?

>>2495624
dumb eighties design - if they aren't bent they short with the microcomputer shielding on top

>>2495632
I'm not so sure about adjusting it while running...

>>2495633
Turn it off, discharge the electrolytic caps and let it cool down for a while. Disconnect the speakers, turn the volume all the way down, power on and measure/adjust the pot.

>>2495636
>discharge the electrolytic caps

Do I need to desolder before discharging them?

>>2495644
No, just short the positive leads to ground with a jumper and screwdriver.

Has anyone mounted an LCD by cutting a hole in a metal case? I suspect it will probably look ugly without some kind of a bezel / sealing /gasket around it to hide the rough / uneven edges unless they are perfectly machined. Is there a DIY-grade generic gasket material or some kind of tape that could fit any screen size by wrapping / tucking it around the screen?

>>2495665
>hide the rough / uneven edges

this is easy if you print an overlay either on thick paper, or acetate.
if you use acetate, you can even include labels or units for the numbers on screen.

>>2495636
>>2495644
Mate they're going to be like 24VDC or 36VDC maximum. No risk of injury whatsoever. Just keep your fingers away from the power transformer and its inputs.

>>2495691
Oh alright, also a guy recommended me a Fluke meter - is it worth buying let's say a Fluke 117?

>>2495702
Depends. Will it earn itself back in a few months? Or even a year or two?

>>2495738
Definitely not haha I'm just a hobbyist trying to get 2 hifi systems going

Do you know of good cheaper ones?

>>2494731
The BMS seems reasonably big. You'll be fine, Anon. Don't worry

>>2495891
ANENG

>>2495891
I just bought a ANENG AN870, it seems like a above average for electric circuits but also not super high end multimeter.
Also most of the cheap fluke's are for electricians / trade's people working with mains, so they apparently don't have measure low values that you would have in electric circuits (but I might be wrong, maybe the super small values are so small that you might not ever need to go that range, I just ordered my first multimeter).
I also contemplated about the kaiweets km601, but the main turn off for me was that the battery life is not great (the features it offers over AN870, was that it has a bar graph at the bottom of the screen, and the lights will blink for which holes you need to use, probably something else I am forgetting about)
You could technically go for a multimeter that's half that price, and it probably does just the same job, and you might even blow it up by accident, and instead of buying a fuse you can just buy a new multimeter since your was only like $20 and maybe you have complaints, or maybe none and you just buy a fuse.

>>2495942
+ I have also seen UT61E+ being well regarded, but it's way out of my price range.

>>2495891
There’s some great Brymen ones (see the EEVblog store), then there’s the Chinese market Fluke, but you’re still looking at hundreds of dollars. Uni-T have been around long enough that I’d probably trust what they put out, and ANENG have a few good meters like the 8008/9. They also have a few shit meters, like the Q1. I’d avoid anything without a rotary switch.

Can I generate RF emissions if I toggle a relay fast enough?

>>2495989
No. It's too slow and it'll burn out.

>>2495991
I've got ones with a max switching time of 10ms and I'm trying to recreate some of this phenomenon https://en.wikipedia.org/wiki/Tempest_(codename)#History which was electric-typewriters and similar devices so I thought something really simplistic could be possible.

>>2495995
100 Hz is pretty low frequency, and the relay contacts have a limited lifetime, usually 100,000 cycles minimum and up to 1 million cycles before failure.

>>2495999
Thanks for answering my silly question anon

>>2496004
You're welcome and it's not a silly question at all. There _is_ an electromagnetic field around the coil that can be detected.

>>2495989
No, but you could use the contacts as a spark-gap. For a few seconds.

>>2496009
There’s a magnetic field, but not much of al electric field.

I want to 3D print some sort of variable guitar pedal body. One such that a potentiometer is coupled linearly to the motion of the pedal. Chances are I’d use a small aluminium box as the main chassis, but with the potentiometer mount being 3D printed. I think using string to couple to the pot would be more robust than trying to include the pot in the hinge, and it gives the option of connecting it to a piece of elastic for the spring-return, instead of trying to incorporate a compression or torsion spring. Does anyone know of any pedals constructed like this? All the ones I can find use gears and torsion or compression springs.

>tried repairing a milwaukee M12 battery
>cut old dead cells out of it
>have to replace the plastic bracket that holds the cells in place next to the pcb
>3d print one after trial and error
>buy three 18650s with pre-spot-welded tabs
>start soldering the cells together
>manage to get them connected to one another around the plastic bracket
>cut the remaining tabs so they'll fit up the holes in the pcb
>cut a piece of silicone amalgamating tape into an insulating barrier with slots for the pads
>spend 10 minutes fumbling with the flimsy pieces of nickel to get them through the slots in the board
>finally get it up there
>solder one tab
>solder the tab on the bottom of the battery
>try to solder the third tab
>soldering iron shorts between the tab and the pad it's meant to connect to
>tab glows orange
>sparks
>melts off
>would have to disassemble the thing entirely to try soldering again
think i'll call it quits on this project, i guess the pcb was the cause of the cells dying all along. that or my desoldering cooked something.

>>2495681
Just slap the overlay on the case around the LCD to hide the ugly edges? Is it OK if the LCD wouldn't be flush with the case surface as it would sit somewhat recessed, probably 1/8 of an inch. Do you have an example picture by chance?

>>2496228
>LCD wouldn't be flush with the case surface

last time i did this was on a bike computer with the LCD back about 1/16'', and it was not noticeable.

>an example picture by chance?

nope, pic is close to the one i made.

I'm building an IoT device (which needs a cell data connection) and the A7670x series (C, E, and SA) look good from a price standpoint but the C & E are listed as intended for Europe, Middle East, Africa, etc. I'm in North America. I already checked a bunch of cell carriers (including mine) and they all support LTE on at least one band that the A7670x chips do. I even have a A7670E running right now getting ~300kb/s on my local network. I know LTE can go much faster but that's fast enough for my application. Am I doing something really stupid if I go ahead and use one of these chips (probably the E but could end up being the C or SA) instead of something specifically marketed for North America?

>>2494282
so i am giving it one more try before just binning the whole project
buying exotic ics from chinks is too risky but i found a popular opamp used for audio shit in a local eshop
Could that work for my microphone preamp?
The main difference is, this it not instrumentation amp but opamp instead (and also 10 times cheaper than most instrumentation amps that aren't chink shits)

the main difference is that op amps have no gain resistor option, so i'm not sure if i can just drop it in as replacement or not
>>2493415 since i would probably have to add a feedback like this right?
op amps also have lower input impedance than instru amps but i think the mic capsule with the tranny should have enough juice to drive it

>>2496391
>>2496392
Simple inverting or non-inverting op-amps aren't properly differential, you'll either want a 3 op-amp instrumentation amplifier (see the diagram in >>2493415) or a single op-amp differential amplifier (pic related). Just find whatever op-amps are specced the best for your application, and ensure the input impedance isn't too much of a problem. In this case I think PSRR should be your main concern, but you should go through the calculations to figure out what the major source of noise will be anyhow.

>>2496402
>>2496402
>>2496402
NEW THREAD