/diy/ - Do It Yourself

>want to design a board with two Trinamic parts
>TMC4671 feeding a TMC6200, just as recommended in their appnotes
>want to put them on the same side so i can use a reflow plate for soldering
>pinouts line up perfectly only when one IC is on the other side of the board
serves me right for trusting the germans to not engage in self flagellation

>>2755576
>serves me right

psychic masochist here, enjoying your bout of psychogenic pain.

>>2755597
i've got to be a masochist at this point, i'm seriously considering buying some 0201s.

>>2755576
Quit being so melodramatic and just stack them.

>>2755510
Posting it again because I did not get any response in previous thread. I am looking for my first oscilloscope and found a great local deal for FNIRSI-1014D. What does /diy/ think about this model? is it worth it?

Is it normal for these bulbs to blacken after only a hour of test use? I rather not tear down my car dashboard again because of these.

>>2755641
What kind of projects are you working on? And what price did you find? That could be way overkill, or it could be less than you need depending.

>>2755708
Small scale circuits, often with 555 timer and lm386, but I also want to get into radio transmitters in the future (I am rather a beginnner). I found the offer for around 130 USD.

>>2755714
Yeah, you will be fine with a $30 portable battery scope. When you start needing something more you can splurge on a good benchtop instead. You'll keep using both because the portable battery powered one will always be so much more convenient, but for some applications you will need something more serious.

>>2755641
this may give you a feel for what that scope is capable of. i think its pretty useful for the price.

https://youtu.be/0h8S36g4zZo

>>2755715
okay, thanks. I guess I will do just that.

>>2755715
>you'll keep using both because the portable battery powered one will always be so much more convenient
Seconding this, it's good to have both

>>2755665
Maybe thats some “getter” that changes the gas in the bulb when it’s used.
If so, it increases the lifespan.

If I replace R1 with an LED, and r2 with a beefy resistor of appropriate value (where it's resistance is current needed x0.6) in mt3608, will it work as a boost, constant current led driver, or shall I expect it to kill the led with random current spikes?

>>2755779
They were dogshit. Burned out in a few hours. And that was at 13v instead of the usual 14.7v the car has. Fuck chyna. Getting my refund now.

Ill be buying 2200k 3mm leds instead and retrofit that instead. Just hope the CRI is high enough to work with the colored filtered lenses. But nothing ever looks right

>>2755641
Sounds like even less of a brand name than Hantek. Look for a review.

>>2755715
Better not be recommending a single-channel scope. They're far less useful than a double-channel.

>>2755787
While that will work, the power wasted on the current sense feedback resistor could be kinda high. I'd consider using a much smaller resistor and a ground-sensing op-amp (or dedicated current sense amplifier) to amplify it for feedback.

>>2755641
It's alright.

I got mine for 115$ with a coupon. You'll learn a lot form it and still be able to resell it for 100$ if you don't destroy it. Mine payed for itself several times over working with the local HAM club.
It's not really good for 100Mhz, more like 30, and if you're using two channels I wouldn't trust it over 15Mhz.
But it's good enough for most of the stuff I do, I mostly use it for HAM HF QRP stuff in the 3.5, 7 and 14Mhz ranges where it's good enough.
It also good for debugging electronics projects, I2C and UART communications etc...

It also has a very rudimentary software defined spectrum analyzer that comes in handy, arbitrary function generator, an okay X-Y mode and you can save images form it to an SD card. This makes it pretty good for matching components, testing filters, checking for first and second order harmonics all the good stuff.

And a thing most people don't really know how to appreciate and auto button. It's really easy to fuck up an oscilloscopes settings when you really don't know what you're doing, and when a lot of things are digital and behind menus it's very hard to unfuckup, so the reset and auto functions helped me out a lot when I was starting out.

Although it's not the best thing in the world nothing can beat it's price point.

The one think I feel it's missing is the ability to connect to a PC and then share the traces. I feel like getting that would make it an incredible scope for any price range. It really depends on what you plan to do with it. If you're crafty you can accomplish quite a lot with it.

You can probably find a better analog oscilloscope on the adds eventually, but the shipping will cost you more then this little thing and you will say bile to a lot of the conveniences it provides, and the score you get will be just as scope and nothing else.

>>2755962
Yeah, double-channel is mandatory for any non-trivial work, but at the stage he's at he'll be fine even with a single-channel to be honest.

>>2755965
Thanks
Kinda non issue for resistor power wastage
Got a 300mA, 20V led(s) cob.
6W on led and 180mW on resistor

>>2756145
Ah, 0.6V? Should be fine. It's mainly the 2.5V references where you have to care about an amplifier.

>>2756037
Usually one’s first project is to make a dual channel to single channel multiplexer box. It kind of switches between the two channels on your mux box into your oscilloscope’s single channel. Kind of like interlacing.
Needs high persistence phosphor though, or you’ll get a headache, like watching 720i TV.

>>2756173
That assumes he's buying a single-channel analogue scope. Which might exist, idk I've never seen one (the chinese actually make them). The single-channel scope anon was recommending a $30 battery scope, like a DSO150.

You could maybe try to do something synchronously, such that every second refresh of the screen is with the other signal, but even ignoring the flicker those scopes tend to have a poor refresh rate.

I kinda want to use a CRT colour TV as a 3-colour 3-channel oscilloscope. Sounds fun.

Have you tried panelizing your PCBs with JLCPCB? Their instructions are confusing. I am not sure if I should just combine everything on one PCB and outline the boundaries for my own reference and then cut myself or I could do proper panelization in Alitum with v-cuts and they won't charge extra for that? How are v-cuts recognized in the Gerber file? Just an outline on a mechanical layer? Not sure what makes them v-cuts.
What do you make of these instructions?
https://jlcpcb.com/help/article/96-pcb-panelization

wtf do you even use a 4chan(nel) scope for

>>2756467
Literally 2chan but x2. Mostly those usecases are better handled by a logic analyzer though.

>>2756467
It’s not often, but I’ve definitely run into situations where I want to look at 3 or 4 analogue traces at once. Like an input signal, output of a transistor, and the power rail to look for ripple. If you have repeating signals it doesn’t matter, you can probe A and B, A and C, B and D, etc. But if you have non-repeating signals you want to catch it all at once, hence why I recommend 2-channel analogue scopes and 4-channel digital storage scopes. You can also have two differential channels by having A-B and C-B, which can definitely be useful.

>>2755510
Where can I learn more about RF parts?
Analog mixers, oscillators, PLLs, synthesizers etc
I'm mostly into IoT devices but I recently bought an rtl-sdr dongle and honestly I barely understand how these work which is unacceptable to me

I tried to build oscillators and

>>2756613
RF is not easy to understand.
Also, what frequency range? AM radio (for example) is not that high a frequency, which is why even the cheap-ass computers back in the day had shielding on them.

I ask questions about RF transmission, reception, antennas, FCC rules, and other things in /ham/

>>2756615
>RF is not easy to understand.
I mean how hard can understand it be? I'm not talking about getting a greybeard level of understand but just enough to figure out what's happening. At least to be able to design things in theory without worrying about practical considerations.
Id like to understand just enough to be able to design modulators and demodulators, and receivers and transmitters out of passive components and BJTs for any given frequency and band (everything below 5GHz, because i hear things get whacky above that)

>>2756613
PLLs are cool, watch a youtube vid or two on that. W2AEW has some neat videos on radio components. The signal path of an SDR is as follows:
>input signal
>amplifier probably
>mixers both being fed the IF, 90° out of phase
>low-pass anti-aliasing filters
>ADC input channels
The important parts to understand are the operation of the mixers, the reason to use 90° out of phase IF signals, the importance of anti-aliasing, and the algorithms used to reconstruct and demodulate the original signal. Fourier is good to know.
Take a look at the HackRF One schematic for more info, though since it can both transmit and receive it's a bit more complex. It's only half-duplex so it could be stranger, unlike the LimeSDR.
Non-SDR circuits may have other features like interesting amplifier topologies and analogue modulation techniques via VCOs or VCAs or whatever. You'll also want to understand transmission lines and impedance matching.

>>2756653
> the signal path of a SDR….
> The Signal Path
Yeah, watch all this autist’s videos and report back.

>>2756648
> 5GHz
Yeah, you’re starting to get into hardline and waveguides and you need a $1M scope.

Also extremely low frequency, used to communicate with submarines is even more wacky.
It’s amazing what you can do with a 37 mile long antenna.

>>2756613
w2aew on youtube
old greybeard from techtronix teaches basic RF concepts, very accessible

>>2755510
Few years ago I saw the ksger t12 shilled for having a good performance/price ratio when just starting out. Are there better alternatives for a beginner nowadays? Mostly planning on fixing broken computer parts.

>>2756767
yihua's, quicko t12's or Weller WLC100 can also work.

i have a thing that needs a 12 volt charger. i have two power cables that both are 12 volts, but the amps are different. does it matter?
i figure if the thing needs like, 1 amp, but the charger does 2 amps, it's not like it will force 2 amps into it, the device will just pull whatever it requires. is that right or do i need to match the amps too?

>>2756887
The amp rating at the source for a constant-voltage source is the maximum current it can provide. If the circuit only requires 0.01 mamps, it will provide only 0.01 mamps, even if it says "2 amps".

>>2756887
>a 12 volt charger
>power cables that both are 12 volts
There is a difference between a 12V charger and a 12V power supply. A 12V charger is something that is directly connected to a 12V battery, often a lead-acid battery, and provides a proper multi-stage charging curve. A power supply just outputs a constant voltage. Many battery-powered devices have a built-in multi-stage battery charging circuit and just want a constant voltage, but not all. You likely just need a power supply, especially if the input says 12V on it. But if it says something like 12.8V, that's a sign that it probably wants a dedicated battery charger. If the old power brick had a light that changed from red to green when it was finished charging, that's also a sign that it wants a battery charger.

I need to replace the pots on one of my guitars. These run a pretty low-tech basic large audio pot, 500K or 250K depending on where in wiring. Mine needs 2 of 3 replaced and will probably just do all of them.
Without getting into the realm of audiophile silliness, how can I ID higher quality pots? Basically low noise, reliable, durable. Is there a particular brand I should look for.
Pic related appear to be on the better end. Note shaft is brass and runs all the way through the case for stability.

>>2756988
>brand
Alps Alpine are good, I think they're the main guys who make open-chassis pots like that, go to their website and browse their catalogue. Though you may want to get closed-chassis pots to prevent dust ingress. Check the datasheet for cycle life, and maybe anything about force tolerance. I'd guess the larger open-chassis pots are more durable than the smaller closed ones, but who can say.

I bet you could use a magneto position sensor as a variable coupling transformer for volume adjustment without any contacts wiping against one another...

>>2754585
> Almost all the THT soldering was done by a machine
We’ll do anything to be able to fire an employee. Even re-roll a board and waste two 1K ballast resistors to give some bear-off to a high current diode because our THT P’nP is so stupid.

>>2757071
…Here’s the other side for doubters.
We have to override the schematic/PCB warnings.

>>2757071
>do anything to be able to fire an employee

a machine is always gonna be better value than whatever human dregs crawl out of the gutter to accept a smelly, manual-labor, minimum-wage job with zero prospects.

>>2757071
>>2757072
hahaha
that's awful

Am I being autistic for wanting to build a harmonic filter for the non linear loads in my room?

Also anyone know what they pay for electronic repair techs working for a utility company?

t. electrician with soldering skills and a 2 yr degree in electronics engineering

I did it, bros. I'm a tester for an electronics assembly company and if I'm good enough they'll move me to soldering. Am I finally gonna make it?

>>2757110
Like $8.50 an hour, no overtime lmao

>>2757161
janny can finally get paid

>>2756988
CTS and Bourns

>>2757072
They're unconnected? Amazing.

>>2757186
>>2757057
Thanks, will check them out
>>2757057
>I bet you could use a magneto position sensor as a variable coupling transformer for volume adjustment without any contacts wiping against one another
The issue is passive vs active pickups. Passive guitar pickups are pretty low tech, Pic related is Stratocaster wiring hookup, there are a million variants that don't vary that much. My Fender bass has something similar. FYI all the "schematics" you'll find look like this, lol.
You can do more with active pickups, but then you need a 9V battery. Dead battery = dead guitar, and I'll tell you from personal experience, they're always dead and you never have a replacement battery when you need it. You'd think they'd make more rechargable stuff, but musicians are luddites and everything seems like its a decade or two out of sync.

>>2757161

wtf?

>>2757275
For some reason, I thought each string would have its own pickup coil and there would be 12 leads coming out and some active filters and mixers inside the guitar body to combine it into one output.

>>2757284
would actually be an interesting construction, the nonlinearity of distortion makes it hard to play harmonies on a single guitar
1 pickup per string -> 1 rat per string = interesting solo possibilities

>>2757284
It looks like pic related. Those cylinders (poles) can also just be iron stuck to a single magnet on the bottom. One pole per string, but just one coil. Works on induction; strings have iron in them as well.
>>2757286
I've built instruments with single coil per string, but hooked them together (parrallel or series, don't recall which), but didn't really experiement with trying to mix them. You can make simple single coils with sewing machine bobbins, strong 1/4" magnet, and #40 magnet wire. They used to sell magnetic toys and I stripped the magnets from those.

>>2757308
> pic related

>>2757308
>>2757309
And here's a Humbucker, which has 2 coils wound opposite direction. These don't pick up 60hz environmenal noise like single wounds, but don't sound as good either (IMHO.)
Seriously, this kind of "musician audio" has a ton of potential for products. Unlike audiophile shit, there's ability to make part that have noticably different audio profiles. (There's still some audiophile silliness but good parts make a noticable change.) And there's a willingness to pay by musicians that want to sound different than everyone else.

>>2757308
>#40 wire
sounds like ass to work with lmao i can already break 30's with my bare hands

>>2757310

>>2757312
It is not easy to work with. The spool needs to be in very good condition so it unwinds cleanly, and you need to get 5-9,000 winds on a coil without breaking.
That said, I had less issues with winding bobbins, than getting thicker wire to solder to it without burning up the #40 wire in the process.

>>2757333
yeh, piezos sound easier to work with. idk why people bother with magnetics desu

Magnetism is fucking complicated.

>>2757333
>put 200 windings on the pickup
>use an off-the-shelf 40:1 transformer to step it up
Problem?

>>2757387
Look up the equations for the force between two magnets, and see if you can calculate what some real-world force will actually be. Quantum tunnelling has easier mathematics than magnetics.

nothing makes my dick feel bigger than needing a multi-channel stereo audio mixer, seeing they're several hundred bucks for COTS units, saying "fuck that", and then assembling one in an afternoon from shit lying around that is as good if not better than said aforementioned COTS units

Oh no not like that bro. What will happen to our beloved suppliers? Ali? LCSC?

>>2755510
rolling for project

>>2757338
Piezo pickups are the go-to for acoustic guitars. I've built several of these as well, using basic round piezos. The output sounds exactly like the guitar in terms of tone. Which ofc is nothing like magnetic pickups. Usually they're paired with a preamp using... a 9V battery.

>>2757585
I don't see any reason why you couldn't just send 9v through the cable to the guitar. what's the current output of a piezo amp?

>>2756767
https://www.eevblog.com/forum/dodgy-technology/dangerous-solder-station-ksger-t12/
Scrap the Hakko T12 chink clones, go with the JBC C245/C210, for example GVM T245 is a good one, 120w 6sec heating time, compatible with the official JBC cartridges, literally holds the temperature (±10C°) even when the soldering iron is sunk into a glass of water, also has proper grounding, you can solder literally anything with it.

>>2757338
Because you can't get proper feedback with piezos. The amp should vibrate the air which vibrates the strings, which the pickup detects. But a piezo picks up the vibration of the air itself, resulting in the high-pitch screech you get from a microphone held too close to a speaker. Feedback is an important contributor to the tone of an electric guitar.
For a similar reason, magnetic pickups are required by e-bows.

>>2757601
Those do look ok, but why do they limit it to 120W? Just have a barrel jack or EC5 or whatever in the back, a software power limit, and let it go the the full 200W that those tips are capable of. Only have a 5A power adapter? Just tell it that. You could choose to use a couple of old ATX PSUs in series, or a laptop brick from the thrift store, or a fresh Mean Well PSU, or even a mains transformer for reliability. Maybe even have a plug for an external thermal sensor, so you don't roast your custom PSU. Not like you'd want to trust the built-in chinky PSU either way, considering their flaunting of proper 240V clearances in the T12 stations. Though in the case of my T12 station, the P-channel power FET that switches the element died first.

These JBC T245 clones are also much more recent than the T12 clones. I wouldn't imagine they'd be much safer than the T12 machines, both in design but especially quality control. That goes for the GVM station and the Aixun T3A.

>>2757601
The ksger machines are just a bit shit and don't have a ground. I DIYed a connection to ground and it was fine.

>>2757595
> output of piezo
Sweet bugger all.
Piezos don’t work well for guitar strings due to their frequency response, and guitar strings are ferromagnetic steel for a reason. (in case you didn’t guess, electromagnetic pickup).
> inb4 “yeahbut I got catgut strings”
The piezoelectric fad, like the hall effect fad, are over. Yeah, it’s “neat-o” but let’s get back to effective means of getting things done.
> inb4 “whatabout LK-99 superconducting strings”
> inb4 “whatabout grapheme strings”

>>2757503

I think anon means like understanding how magnetic flux works and all the theory rules behind magnetism.

>>2757616
Best to connect to ground via a 1M resistor for ESD protection, you don't want a low-impedance line to ground since that can allow very high currents to flow. A floating iron also lets you work on live equipment.

>>2757617
What about electrostatic pickups? Just the same as a condenser microphone, you hold the string itself to a potential (e.g. 50V) and the capacitive plate beneath it in the body of the guitar has a voltage induced in it as the strings move closer and further. You can also go the RF condenser microphone method by making the string-pickup capacitor be one element in a high frequency RC/LC oscillator, such that the frequency modulation can be demodulated into audio via a PLL. The electrical burns on your fingers are a feature.

>>2757617
>LK-99 superconducting strings
Keked

>>2757662
The most fascinating thing is how electret condenser membranes are made. They hold plastic in a strong electric field and let it cool and what’s left is a membrane with a permanent electric field. Just like how a permanent magnet is made.

I’m pretty sure the back-to-the-future flying skateboard can be made with these…

>>2755510
Hello /ohm/ ! Came here from /ham/ since I still have some unanswered questions. I am trying to build simple FM transmitter according to pic related. L1 is just standard choke with 10uh and L2 is 0.4mm copper wire wrapped 6 times around 2mm nail. Here is the video with working transmitter.
https://www.youtube.com/watch?v=YSrp4l74yvE [Embed]
I have tried to use BC548 and make both coils myself but I havent had any actual successes. my crystal has 4 pins and looks like this (I think that both middle pins are just GND)
https://imgur.com/a/0bB9YhI
I am pretty new to electronics (and very, very new to radio overall) so I apologise in advance if I am doing some simple stupid mistake. My main questions are:
- Can I just use 2 axial chokes instead of 1 choke and a coil? If so, can they both just be 10uH?
- Is there any decent tht equivalent to BFR93 or should I really use this transistor?
- Is my crystal good enough or should I get another one? /ham/ suggested that there might be something wrong with it.
I am mounting it the same way as this guy does in this video and powering it from external DC power supply. I monitor the signal using RTL-SDR. If there are any things I should change, It would be nice if you could point it out. Thanks in advance /diy/ people!

>>2757617
> piezo
Don't know what to tell you. I can tell you it works from personal experience, and that kits for acoustic guitars include piezo bridge pickups, pic related. Guy I'm playing with now is using one.

>>2757695
>L2 is 0.4mm copper wire wrapped 6 times around 2mm nail
Is the wire insulated?

>>2757662
Like the metal chassis of the power/control box wasn't grounded at all. It's a 3 pin iec plug but the third pin isn't connected to anything. The other two were soldered directly to the board even though they weren't the right type of connector for that

>>2755787
your design idea won't work.
r1 and r2 form a voltage divider such that the voltage at fb will be some scaled version of vout.
an led will always drop 0.6 or 0.7 V.
this means that the sensed voltage at fb will no longer be a scaled version of vout, but rather vout-0.6 V or vout-0.7 V.

>>2757695
>my crystal has 4 pins

that could be a problem.
it could be a powered oscillator, which wont work at all in your setup, and cant really be adapted to work with it.
also, your transistor is rated for 150Mhz, while the ruskie one is 6Ghz. big difference.
can you use 2 chokes? no, you need L2 to oscillate at RF frequencies. the job of a choke is to block RF.

>>2757802
My crystal looks slightly different (see the imgur link in my original post) but it still might be that I need other one, so I will probably need to order those. I think the other problem now might be the wire I used to make the L2 coil was not isolated, just like >>2757724 suggested . I now feel very bad right now for making such an elementary mistake. I guess I will just order a transistor with proper frequency and get better coil. Again, thank you everyone for help! I will probably let you know eventually if I get this circuit to work.

>>2757807
>I now feel very bad right now for making such an elementary mistake.
You started off by saying:
>I am pretty new to electronics (and very, very new to radio overall) so I apologise in advance if I am doing some simple stupid mistake.
It's the natural order of things, Anon. You just learned something new and increased your bomb defusing skills by 2 points. Be happy.

>>2757811
thanks!

>>2755510
Im trying to identify this diode.
computer is not powering on and its burst (diode is near ac adapter port)

its 4mm x 3mm, surface mount
it has "KW" on it. its black, and theres a blacker line underneath the letters.

>>2757969
See:
https://smd.yooneed.one/code4b57.html
The diodes are the parts starting with DO, there’s a few diodes in SOD-123, SOD-323, and DO-214AA. Measure your diode to see which package it is, and hence narrow it down. You can also make some educated guesses about the likely manufacturer, by looking at other manufacturers and seeing if they’re mostly american, chinese, euro, jap, etc. You can probably also guess the voltage rating and type of diode based off the context of the circuit it’s a part of. Especially if it’s right next to an IC that you can look up the datasheet for. It’s probably a ~1-4A Schottky diode, but it might be a zener, or maybe a normal PN diode.

>>2757969
Forgot macro pic?
Anything on the silkscreen on either side of the board? … like “ZD4” or something?
Are there 3 other diodes in close proximity?
Is there a test point callout with something like 5V1 or 5.1 somewhere close by?
Is it right by a coil or inductor?

i have aPH laptop battery stuck in "permanent failure" mode after being rebuilt, the chip in it is BQ2084, is there any way to reset it with a special magic song and dance over the I2C data bus that is exposed on the connector?

>>2758053
HP*

>>2758053
Can't see anything about it in the datasheet. Maybe you can clear the Permanent Failure Flag register, but I don't think so. I'd probably just order a replacement IC, they're under $10.

>>2758053
Those bms chips make it so hard to build batteries for old PDAs and those sort of gadgets.

>>2758053
I’ve had some luck resetting them with shorting some pins out (accidentally) but this is also likely to kill it. It was random.

I’m pretty sure the EU should be working on banning those practices.

>>2758089
they only say this in there
>The SAFE output and PF Flag register can only be cleared using a series of ManufacturerAccess( ) commands.
i wonder if ti has ever had software that may contained this set of magic commands, and writing 0x00 to address 0x11e does nothing as its a read only value

>>2758112
nvm the related software is written in visual basic 6, can never decipher anything from this junk

>>2758112
Good luck. They spent more time, effort and money implementing the DRM shit than they did with the battery management stuff. The DRM and related support is probably 90% of the die.

>>2758144
where did you find that? is there code available or am i going to need ti implement SHA1 correctly on the Arduino?

>>2758214
sha 1 and whatever the factory unlock key is, but might be changeable so id probly have better luck trying all 2^128 keys

>>2758226
>2^128
Too big to bruteforce.

>>2758286
that's what she said
did it anyway

>>2758144
>all this protection
>for a fucking battery management IC that only 3 people are ever going to bother to reset or reconfigure

>>2756648
Things get wacky in the hundreds of megahertz range very quickly as well. You do not need to understand RF to understand modulator and demodulator, you only need to understand the theory of digital communications.

>>2757724
>>2757807
topkek, I would have never been able to come up with this, you really had low expectations from that guy didn't you haha.

>>2758019
It is an apple powerbook from 1992. im guessing they used high quality japanese diodes. I would be out of my league to guess which .
>>2758038
Here is a pic of the bad diode (tiny one) and the neighboring components. The three soldered legs to the right of the pic - on the underside is the power plug, which takes in 7v 2ma

>>2758477
So, I think you mean D15. Just by the look, I’m guessing this something like a 1N5404.
However, in a switching power supply you probably want a Schottky. If you measure the AC and DC voltage across it and it’s low (under 40 V) you might want a 1N5822…
Or their surface mount equivalents.
Those are 3 A diodes, you might want to go with higher amperages, like 5 A, or higher voltages to be on the safe side.
Also, something else might have failed that caused that diode to go out in the first place, when you remove it, probe around to make sure the diode isn’t in a shorted current path.

>>2758477
>7v 2ma
huh?

Anyhow, measure it to figure out what package it is. Judging by the fact it's half the width of a 6mm wide DPAK, it's about the size of an SMB diode. All of those I see in the codebook there are TVS diodes with a breakdown voltage from 9 to 14V, suggesting this one died due to a spike. I'd check if its cathode (banded side) is on the same node as any ICs' positive supply pin, and see what voltage is likely to be on that supply.

Don't suppose someone has posted a schematic for that computer? If it happens to be an aluminium powerbook G4 I happen to have one to probe at, not that I think it will turn on considering the state of its charge port and battery.

>>2758497
Cancel this
>>2758516
Yeah, it also looks like it could be a TVS diode. Totally different.

>>2758497
yup, d15. w/o power its measuring 1.227 k ohms across it.
I cant see anywhere else on the board that looks bad, all the points in that circuit area look good.
im not sure how to check for that diode being in a shorted circuit path. heres the back of the board. those large (caps?) are opposite the bad diode

>>2758520
This guy >>2758516 is probably right, a TVS diode. These protect from overvoltage. Someone must have plugged it into a massively powered, over-voltage source I reckon.

>>2758520
L2 and L3 (L12 too) are inductors or chokes. Considering the LT charge pump ICs and op-amp IC, they're likely smoothing out voltage rails for analogue circuitry. Inductors and chokes produce voltage spikes out of sudden changes in current, so having a TVS diode nearby makes sense. Is it near speakers or a headphone output?

Also check that those tantalum caps (C1, C39, etc.) aren't shorted out too, they can go bad over time, and they don't handle transients well. Though usually when they fail they show it.

1.227kΩ is kinda high for a short circuit that would prevent things from working properly, but knowing diodes it could have a nonlinear failure curve instead of just dying into a resistor (i.e. it probably has a significantly lower resistance at higher voltages).

It should be possible to turn the thing on with the TVS diode desoldered, and to measure the voltage across its pins. Then you'd find a TVS diode with a maximum standoff voltage that's a little bit higher, and a maximum clamping voltage that's hopefully below the maximum voltage of any components on the same node. In reality this clamping voltage is measured at like 30A, so you can probably go a bit lower. The Vishay SMBJ10 has a 10V standoff voltage and 18V maximum clamping voltage, while I see 16 tantalum capacitors nearby. Assuming it's a 7V power input, they'd probably be fine. Those are a couple of assumptions though. Wanna look up pictures of the candidate diodes on LCSC and start font comparisons?

>>2758522
If that's the case, it probably wouldn't have protected anything. That's the kind of situation you need a crowbar circuit for.

thanks for all your guys help with the diode thing. I think youve lead down the right path to replace it.

>>2758522
yeah, its a standard 5mm~ plug, and its a laptop. considering it only wants 7v 2kma, someone probably stuck something much bigger in there, trying to get it to work.

>>2755510
>>2755576
Idiot here
I found a vintage car battery charger in the trash, and it seems to be made well

The problem is, it says it's putting out a 60hz signal on the 12v dc out. The AC out voltage is fluctuating alot, seems to be about 5v ac alongside the 14vdc, but will spike as high as 15vac. It's only for charging 6v and 12v lead acid batteries.
I put a full bridge rectifier on the output and the multimeter stopped saying there was a 60hz output. I'm thinking the rectifiers were blown or maybe its just designed bad. I funny really know what these disc things are or if I'm overreacting to the presence of ac, but I think this would destroy my batteries.
It's built with a silicone controlled rectifier and two giant discs instead of diodes, I remember hearing these discs were how rectification was done in the 1800s.

Should I just gut everything but the transformer and drive this with a full bridge rectifier and a triac or what? Or is this silicon controlled rectifier a superior choice for a battery charger

>Friend sends me a link to some startup that auto routes PCBs
>Feed it a simple file that was 2 layers and 76 vias when routed by a human
>This is its "Best" attempt after a half hour
nvm the edge cuts I had to make it a lot bigger because there's a check for pin surface area vs board surface area that fails otherwise

>>2758752
>6 layers
Do they sell the boards too, LMAO

>>2758748
Oh also should I not use a triac on mains to control the voltage out of a transformer?
I used a router speed controller to adjust the voltage but it started saying it was 250 hz. I could adjust the dc out like that to the hundreds of a volt, but I don't know if That's a bad thing to actually do and don't want to burn up this 80 year old transformer.
The overall design seems kinda outdated. The voltage out seems to varry with the mains voltage, and here in this dystopia it fluctuates between 104v-125v over the span of a minute every day.

>>2758768
>don't know if That's a bad thing

no, that's fine.
if it's good enough for a router....

>voltage out seems to varry with the mains voltage

it'd cost twice as much to make it regulated.
battery dont care: sometimes it's charging a lil faster, sometimes a lil slower.
as with jerking off, fast or slow makes no diff. in the end, you'll get there.

How do they make chess pieces identified correctly on a square? They are not using 64 different NFC coils for each square and 64 different triggers in each piece right?

>>2758748
Those are old selenium rectifiers.
They are shit, and will be done after a single accidental short circuit on the output because they are so old.
They were even shit when new, but for battery charging, where even pulsating DC is fine, they suffice.
Replace with high-current shottky rectifiers, which have a lower voltage drop like the selenium rectifiers.
Or, if you want to go full-nerd-gasam, se an “active rectifier” that uses mosfets and a controller that senses the incoming AC and turns on the appropriate mosfets and presents a milli-ohm resistance instead of a .3 or .7 V voltage drop that the rectifiers will produce.

>>2758822
There is a chip and a capacitor connected to the coil. It gives a pulse, and simultaneously data, that powers the chip just long enough to respond.
Well, that’s how a lot of rfids work.

>>2758822
The pieces start from a fixed position and the computer records each movement in turn.

>>2758829
Each piece has a magnet in the base. White = north, black = south, or vice-versa

>>2758819
>no, that's fine.
>if it's good enough for a router....
Idk if you're being mean or if I should believe this
It's not a router it's a transformer

>it'd cost twice as much to make it regulated.
>battery dont care
I hoarded trash to make stuff, my wallet doesn't care cause it's absent.
Isn't regulated better?

>>2758825
So I should just bypass them and use this full bridge rectifier I pulled from a TV with a propane torch?
It's got a fat heatsink, solder gun and hot air gin wouldn't touch it, warped the PCB though.

Do I want to ditch this 1950s GE Silicone controlled rectifier and instead use a triac or mosfet?

>>2758833
>It's not a router it's a transformer

right, which is a much simpler machine, thus more rugged and harder to kill.

>>2758841
Oh, ok thanks
Makes sense being a giant electromagnetically inductor

I usually wouldn't care but it's actually pretty nice. It's half the size of a mot, completely coated in sap, bolted together, and insulated with papar

>>2758748
>vintage car battery charger
I've seen a battery charger that was literally just a transformer and a rectifier, no current limiting or filtration capacitors. That one did also have a thermal switch to prevent overheating. Transformers intrinsically limit current, I've had mains transformer-based power bricks output 12V open circuit but go down to 9V with any kind of load on them, so this is likely what they do to prevent the battery charger from blowing a fuse when you connect its 14V output up to a 10V battery. Lead acid batteries are also pretty resistant to overcharging unless you drive like >16V into them for hours on end even after they're fully charged.

>>2758768
It's considered bad practice to use a TRIAC on the input of a transformer, it results in dV/dt spikes that the transformer can't easily replicate.

>>2758822
Could do it with an analogue hall sensor beneath each pad, with different magnetic field strengths indicating different pieces. North poles would be white and south poles would be black, for example. You probably wouldn't even need to track which piece is which type, considering it could just remember the board state and figure out the only possible piece something could be. So you could really just do it with hall switches or reed switches, though that would be less robust against people fucking about.

>>2758845
>It's considered bad practice to use a TRIAC on the input of a transforme
Is there something that can be used instead, basically want to niggerrig a variac
Also what are the implications of "dV/dt spikes that the transformer can't easily replicate."
Thank you

>>2758845
>ve seen a battery charger that was literally just a transformer and a rectifier, no current lim
This one is just a transformer, src, and two mystery discs. Someone called it selenium, but that says 1930s and what I read said it was 1800s
There's no thermal overload. The font on it is cursive. I place it between 1940 and 1969

>>2758850
Boomer here, I have one from Canadian Tire… from the 60’s or 70’s, I don’t remember exactly. It uses selenium rectifiers.

They are better for low quality, high current devices. I suspect they still had warehouses of these rectifiers they couldn’t sell so they just put them into low-accuracy, brute-force type of equipment like car chargers and arc welders.

>>2758845
> It's considered bad practice to use a TRIAC on the input of a transformer, it results in dV/dt spikes that the transformer can't easily replicate.
I’m not sure what this means exactly, but I would think a triac on the transformer input would act—to some degree—like a filter and snub-out some of the spikes caused by the triac switching so it would be less prevalent on the output.

>>2758858
>Boomer here,
It's an honor to be in your presence sir
Truly the greatest generation
o7 glory to Israel sir

>>2758859
… also, I have put light dimmers (that use triacs) on wall warts and it kills the rectifier diodes every single time. Of course, if they used 1N4007s instead of 1N4001s it probably would have been fine. So some spikes are definitely getting through.
There are filtered dimmers where this might be less of an issue, or make your own filter.
I assume you could make one with fixed component values, but a class D amplifier is able to turn squarewaves into music, so maybe there is hope.

>>2758850
>Also what are the implications of "dV/dt spikes that the transformer can't easily replicate."
I think you'll get more heat dissipated in your transformer. Additionally, since it chops up the waveform, it isn't a very good method for varying the output voltage. But if you have the thermal leeway and want choppy voltage (which battery chargers sometimes do) there's no particular reason not to. My question would be what do you want this TRIAC to achieve? To vary an output voltage with feedback? Because I'd use a buck converter after the transformer if that's your goal.

>src
Do you mean SCR? Where is it in the circuit? Draw a diagram if you can.

>>2758859
Well normally a transformer has a rectifier and filter on the other end anyhow. So using a phase-fired dimmer (above 50% duty-cycle) will not significantly change the output rectified voltage but rather increase the magnitude of current spikes. If you reduce the duty-cycle below 50%, then you can substantially reduce the output voltage, but your power factor is awful and you'll be putting a lot of stress on your rectifier(s) and transformer with the current spikes. In the case that there aren't any filter caps, you will get a lower duty-cycle on the output and the current spikes won't be quite as big, maybe that's a good thing for a battery charger, idk.

>>2758862
> glory to isreal
I said “boomer”, not “retard” — believe it or not, all boomers don’t live in 1976 like biden does, nor thinks like he does.

I’m even willing to accept the possibility that newfangled class D amplifiers might possibly work and aren’t evil witchcraft and voodoo.

>>2758865
>My question would be what do you want this TRIAC to
My mains is nightmare fuel, I just kinda wanted a niggerrrigged variac but also I wanted to regulate the output so I can maintain optimal voltage for the battery, but I'd probably do that on the output. Honestly I was just drunk and trying to test this thing and needed to know if I was hurting it because it was really convenient to change the voltage this way

>>2758865
>Do you mean SCR?
silicone controlled rectifier.
The whole charger is a bolted together transformer, scr, some crap that looks like diods, 2 selenium rectifiers I guess, a switch for 6/12v, and an ammeter
I don't have a schematic, That's why I'm tempted to gut the thing. I found it in the trash, it's a six amp charger, I found a 15 amp full bridge rectifier in a broken TV, and I don't know if I want to destroy vintage equipment that night be fixable. Idk what scr is. It might be better. The silicone on the package feels like plastic

>>2758870
Kek, sorry

>>2758872
>silicone controlled rectifier.
SCR = silicon controlled rectifier
maybe you were getting the letters around the wrong way
also "silicone" refers to the elastomer, not the semiconductor element

>I don't have a schematic, That's why I'm tempted to gut the thing
It's like 6 components, if you can't follow the wires and draw one I don't think you have any chance at fixing or modifying it.

>>2758872
Why are you trying to describe it? Here is mine.
Those two plates near the back are the selenium rectifiers.
The 6/12 V switch works on the primary due to the lower currents (cheaper switch).

>>2758900
… and for those of you who think inflation is bad, here’s what I paid for it back then.
If you have trouble posting images, upgrade to a 9600 baud modem before logging into this BBS.

>>2758901
>install 9600 bps modem in my nicotine-stained MEGA-AT tower containing a 386sx processor running at a blistering 20MHz, drawing approximately 4000W at idle.
>click picture
>suddenly remember the old ways
>what is the resolution of this image again?
>3 days later...
The answer to life, the universe, and everything is $48.95.

>>2758883
I can schematic it, but I didn't see the wiring I expected and haven't dealt with scr so I needed to ask if this is worth fucking with.

>>2758900
I don't know why I bothered. I guess to convey there's no controlboard like new ones have
Mine looks about like that but decades older and I think your transformer is bigger

Just reverse engineered the router module's PCB from my Snapmaker Original. Only P1, P2, R9, and R13 are populated, the rest aren't present. I think that 4-pin IC U3 is an optocoupler, at least it doesn't look like a bridge rectifier. The Q2 path is strange though, I think it pulls down to PWM instead of to GND because the motor uses the PWM node as its negative most rail. The motor sends a signal to Q2, which activates U3, which sends a signal to what I'm guessing is an MCU. Maybe the opto was necessary for brushed motor noise reasons? The MCU can also read the PWM pin. Then the MCU could send info back to the Snapmaker itself by pulling down on the identification pin, which is normally just used with an ADC to measure what size resistor it has and hence to figure out which module is connected. The ID pin is also used by the 3D printing module to measure the temperature of the thermistor. P3 is probably for a fan. Since the MCU cannot do anything to talk to the motor or PWM it directly, I can only assume it was intended for motor diagnostics. Maybe the motor would measure its own current to detect a stall, sending a fault code back. But why the 20-pin MCU then, instead of a direct line back to the mainboard? Maybe it's something other than an MCU, idk.

The motor likely isn't intended to be PWMed, even though it's on a MOSFET that's also used to control the laser module. But now I have to PWM it, since I replaced the old 24V "smart" brushed motor with a 12V brushed motor. I'll fudge it and make it think it's looking at a laser module instead, by swapping that 100Ω resistor for a 130Ω.

>>2758914
It’s worth converting to a 12 V linear supply and/or fixing it. I had mine out and open for the last 5 years to fix it. I was looking for a cheap way to replace the selenium rectifiers.

One thing I found out is that you can’t just put 10 1N4001 rectifiers in parallel and expect that to work like in resistors (they blow up). Nobody really knows what happens when you put diodes in parallel, it’s an active scientific investigation.

>>2758924
>Nobody really knows what happens when you put diodes in parallel, it’s an active scientific investigation.
What do you mean?

Mines a 6 amp model, I found a 15 amp full bridge rectifier 4 pin flat ic bolted to a heatsink. I was just going to put that in there but I can't bring myself to destroy the selenium. I don't know if I should replace this scr too or not. It just feels wrong to gut something old and use it's corpse as a housing for a modern design. Some zoomies might get it someday and get the wrong ideas, but then again it sounds like this design and selenium were trash anyway.
I might just leave it all in there and bypass it instead of throwing it away.

>>2758924
>Nobody really knows what happens when you put diodes in parallel, it’s an active scientific investigation.
yes, we do.
they act like a single bigger diode.
same amount of voltage dropped, twice as much forward and reverse currents.
kvl and kcl still apply.

>>2758988
>they act like a single bigger diode
but not with the same current carrying capacity and thermal specs
their forward voltage has a negative tempco, hence they'll get out of balance without thermal coupling and in the worst case they will die from thermal runaway. you need balance/ballast resistors to prevent this, or just to give yourself a lot of thermal/current leeway.
same thing happens with bjts in parallel, and mosfets in part of their linear region, but not mosfets that are fully turned on.

the anon talking about "it's an active scientific investigation" was flippantly saying "it's more complicated than my simplistic initial thought", in case that has to be pointed out

Is the matte finish on casings a sign of rebranding?
I got some IRF3205s from the chinks and they're all matte on the front.
A quick test with the LCR-T4 shows the same Vt as some original 3205s, but the capacitance is almost double.

>>2758900
Does yours have a lightbulb or something in the transformer?
I can't read it, but it looks like a little light bulb. I accidentally shorted the clamps and it looked like sparking came from inside, was that the lightbulb?

I recall some anon here griping about 9V batteries. My guitarist uses 9V on his and it's constantly dead. Went looking for 9V li-ion battery pod I could wire in and convert it to something that could be charged, and found these.
This is like 10X power density of alkaline, and charger is built in (no need to even remove battery from service to charge.) This is less than they charge for 9V alkaline locally. Had no idea this was even a thing until now.

>>2759040
Yeah we've come a long way. you can get batteries in almost every form-factor that are rechargeable with a little USBc or Microusb port on the side. They're even better than traditional NiMH rechargeable because the old school rechargeable ones that aren't lithium are often lower voltage than normal batteries. they are especially useful for those expensive as shit 123 camera batteries. (picrel) and other sizes that are only really used for one thing.

>>2759040
I wouldn't believe the mAh ratings on these no-name chinese batteries. I would estimate that 9V battery has less than 10% of the mAh it claims.

I know the answer will probably be no, but is there any possible way I can fix this display? The digitiser & backlight are still working all over, but the LCD has stopped working in the right 1/3 of the screen. The dark area occasionally lights up grey/white. it's split into two exactly equal stripes. The total area is 86mm wide, two 43mm strips, and the display is an exact multiple of this.

I'm pretty confident is an internal connector/cable that's the issue, because of how regular it is. Does anyone have any ideas about where to look? I've reseated the LVDS ribbon but it's done nothing.

Here is a video of it:
https://imgur.com/a/Bz0xEfU

>>2759072
If the panel isn't crack it might be, if you're dedicated enough
Sounds like one of the cables went bad or the drivers failed. Real fucking neat how it's just 2/3 shut off.

Get a generic driver or a water damaged donor laptop

>>2759072
FPC ribbon on the panel.

>>2759079
Maybe the mezzanine connector or its solder joints.

>>2759063
Agree, but 10% would still be more than a traditional alkaline. I'd be more concerned with internal resistance killing the battery over a couple days, but won't know until you try it.
>>2759049
>those expensive as shit 123 camera batteries
Ooo. I've several SLRs that run those 123s. They are a pia to find now.

>>2759016
Probably a fuse.
When mine shorts the the rectifiers spark out from the plates!

>>2758997
> diodes in parallel
Heh, something like that. I’ve never seen a design with parallel rectifier diodes. If you had a box of 1000 you might be able to go through and match them and get the desired parallel effect (without ballast resistors) but I’m not that guy.
Plus, all the reverse leakages also add up, etc.

>>2759040
Lol, 16.8 Ah. Not bloody likely. Not legit.

>>2759049
EBL is legit as far as I can tell. One of the few places I can still get NiCds.

>>2759072
Feels like the wide displays might be implemented ith 3 driver chips, and one went bad or something. Seems odd that it’s an odd number.

It's not parallel diodes, but it does something.

>>2759077
>>2759079
>>2759080
Thanks guys.

I think I found the problem: There's a wide ribbon that seems to be an internal connector in the display. Seems I cut through it when removing the screen....

>>2759099
Yep, that'll do it. Take it outside and shoot it with a 12 gauge.

>>2759100
It's *only* about 40 lines, what if......

>>2759102
On a scale from 1-11, how much do you hate yourself? Followup question: are you Korean?

>>2759105
Probably a solid 8, so I can't rule anything out. Not korean or I'd have had it done by now. Got any tips? I've got some enamelled 0.4mm wire but it seems way too big. Is enamelled necessary or can you just use individual strands from a bigger wire?

>>2759111
If they're not enameled, how are you going to prevent them from shorting on things?

>>2759111
You have mustard under your thumbnail, but that's not important. What's important is the fact that what you want to do isn't worth the trouble because you're not Korean, and you don't have the tools to de-laminate the panel glass and replace the FPC ribbon before re-laminating the glass (no bubbles or cracks allowed). Might as well just kill yourself now.
RIP, Anon.

>>2759113
I was thinking of using some flexible rubber compound over the top, maybe UV curing or something.

>>2759115
It's from peeling an orange but point taken. You gonna come check for bubbles are you? huh? Gonna give me a mark out of ten? Yeah I think it's probably dead. I'll keep my eyes peeled for a donor tablet I can take the display from. I hate how disposable this stuff is. Glued displays & batteries etc should be illegal. RIP me indeed

>>2759123
>I was thinking of using some flexible rubber compound over the top, maybe UV curing or something.
>I don't want to use my enameled wire, so instead I want to take normal wire and enamel it instead
Just buy smaller gauge enamel wire

>>2759123
OK call me mental but I've just ordered some 0.16mm enamelled wire

>>2759125
Yeah fair, just thinking of using what I already had (my 0.4mm enamelled wire is far too thick)

>>2759094
>When mine shorts the the rectifiers spark out from the plates!
Then that'd probably what it is.

I think I'm just gonna leave the old crap in there for a gen c kid to find, and shove this 15a tv full bridge rectifier in there and call it a day.
Thanks again

>>2759111
I don't think that'd supposed to look like that.
It appears to be a zebra cable thats both unglued and ripped in half

>>2759123
>It's from peeling an orange but
Because citric acid is good for pcbs and adhesives, right?

>>2759098
Can I put my dick in it?

>>2759149
kek this is the coolest thread.

>>2759149
As a matter of fact, you can. You ever heard of a TENS machine? This is a THOUSANDS machine and it grew my benis 2mm in 14 days. If you decide to take up this benis exercise regimen, you should shave first and use electrode gel so it doesn't burn so much.

hello, retard here.
so currently have one of these L298N units connected to an Arduino Uno and a very silly battery mess.
to my understanding this should be correct, and it has been working fine for a while, and then suddenly it stopped working, as in the indicator light is on, but the output does not power the motor.
i replace the L298N with another one, and it behaves the same.
with a multimeter, while the 5V is off, the Vs terminals read the expected ~16V from battery. with 5V on, these drop to around 1V. the output terminals stay 0.
here's the weird part, when i measure the batteries at the battery terminals while 5V is on, the whole thing starts working again and the motor runs as expected.
please explain. why does reading the battery terminals make it work? have i broken this L298N unit in a retarded way? did i immediately break the second one in the exact same way? why did it work fine before?
thanks for any help you can provide.

>>2759325
have you reflashed the program on the arduino?

>>2759168
This anon knows what's up. Also, an electrode in the pooper makes it better.

>>2759350
yes. I have two Arduinos and i flashed/reflashed the program to both of them and tried both.
It is a very small program that takes the input from a switch(thick white wires) and uses that state to determine the switch pins (yellow and teal).
In the image i have just taken the switch pins out of i/o and made one 5V, the other GND. this was to simplify and test and the behavior holds

>>2759351
No he doesn't.
Tens units don't use kvs, that would destroy cells and cause DNA damage and burns. They use high frequency low voltage.

>>2759325
You need to disconnect the 5V from the L298. Just connect In1 and In2 to the arduino.
Also reattach the jumper.

>>2759355
i was under the impression that that jumper connects the Vs to the 5V regulator to supply the 5V logic power. My Vs is greater than 12V and so would be dangerous to the regulator? 78m05?

>>2759357
>>2759355
tryied it anyway because pretty sure regulator has thermal shutoff.
behavior is the same, motor runs only when multimeter leads are touching battery terminals.
this time indicator light only turns on when motor runs (leads touching terminals)
multimeter can be on or off btw

>>2759359
Well in exactly this setup it works for me. Only difference is that I have a 9V power supply and am not using 9V batteries.

Looks like I have the same pcb and am also using an Uno.

>>2759359
Sounds like your batteries are dying, and the multimeter is set to measure ohms, and puts a 3rd 9V battery in parallel.

Stop that battery nonsense. You don’t use linear regulators with battery circuits.

>>2759354
>Tens units don't use kvs

nonsense.
in order to activate a muscle from outside the skin, you need 100-200 volt pulses.
i have several units, including one i designed and built.

>>2759359
>multimeter can be on or off btw

sounds like your battery solder connections are intermittent.
flex/push/press on them with a pencil, fingers, whatever, to see if that fixes it.

>>2759378
> soldered onto 9 v batteries
Omfg…

>>2759367
>>2759378
>>2759380
thanks you guys. just checked again, was definitely the batteries. my rarted decision to solder directly onto the terminals has most probably destroyed one of the batteries as evidenced by the melting plastic about said terminal. it wasn't that the multimeter was doing anything, but pushing on the terminal was causing the internals of the battery to function
>>2759367
this project will eventually need to be battery powered. i am not using the regulator on the L298N, but will eventually need to regulate the ~12 to 24V battery that i imagine i will need for the some of the motors down to Arduino 5V and 5-7V for an additional servo motor. what do you recommend for battery powered motorized applications? buck converter?

>>2759386
>solder directly onto the terminals

you can do it if you do it smart.
trick is to use a file to remove a very small section of zinc/nickel coating on a corner section.
key words are SMALL and CORNER.
then put a small blob of solder there, tin your wire and attach.
(for AA batteries, remove some of the plastic sleeve to reveal the corner on negative terminal.)

>>2759377
>150V != kilovolts
150V is pretty high, the usual values I see are 50V

>>2759386
Yeah, I'd use a step-down converter to power the microcontroller. If you're not a retard, the MCU power draw will be so low that it won't make a significant impact on battery life to just use a linear regulator. But if you're dropping it to 5V that you can also use for the servo, then yeah a buck converter makes most sense. If quiescent current matters, there may be ways of having the switching converter turned off until its needed. Well if quiescent current matters, you wouldn't be using an L298N and an arduino operating at 5V with a USB-serial bridge, not to mention all the LEDs.

What's the project?

>>2759394
>the usual values I see are 50V

you been looking in all the wrong places.

>>2759393
>>2759394
thanks for the tips. its an animatronic wing cosplay prop

>>2759403
I do plan on making a TENS-like system in the future, I want to make a large dot matrix of pixels laid across the skin, where the intensity at any one pixel can be independently varied. In an Applied Science video, he makes an electroluminescent driver for a dot-matrix EL display, which seems like it would sorta be applicable, but it lacks a high-z mode, which I think is the reason for the pixel intensities not being fully independent.

But that table suggests that current-based control is important. Would just putting a resistor in series with all the columns or pixels give me a better result? Also do you know if there's anything wrong with using a square wave as opposed to a sine wave? I know that you want the net DC voltage to be zero, which I believe Applied Science addresses for his EL driver, but I'm not sure about the other engineering guidelines for such a project.

>>2759386
I suspect, but can’t prove, that step down buck converters are going to be more efficient (generally speaking) than a boost converter.

>>2759420
>current-based control is important

it's a safety precaution, so its only important if you're risk-adverse (or a lil girl).
a small shock becomes a big shock if the skin is broken and current isnt limited.
that could cause someone to jerk away and hit you in the eye.

>>2759420
>using a square wave as opposed to a sine wave?

typically you use pulses in the microseconds. 100us is what i aim for.
like the pic except dual polarity.

>>2759377
I've put a multimeter on one and it said it was like 20v ac

>>2759473
>multimeter

dont ever believe what you see on a multimeter.
scope it and note the very low duty-cycle pulses at high voltage.

Does a VFD for big-ish motors have to be sine wave?
Whats stopping me from taking a cheap brushless DC motor controller and hooking it to three solid state relays to get a 25amp VFD for cheap?
Will that kill the motor in a hurry or is three phase more resilient to non-sine waveforms?
(I realize I will also need AC-DC conversion before the relays)

>>2759359
The 5V regulator on a L298N board handles 24V for a few minutes. After that the board needs 5V to be supplied.

Whats a good starter multi-meter? I'm looking to maybe so some mod-chip installs on PS2s and xboxs so i'm thinking the bulk of the use would literally just be checking continuity... but if i grow into the hobby more, better capabilities would be good. A lot of the research i have seen points to the AN8008/8009, and then maybe swapping the test probes with something thats a bit better with a gold coating on the tips.
...but most of those posts and discussions are from 5+ years ago.

>>2759494
The AN8008/9 is fine for what you're doing. Expensive meters are more about certification and calibration, and aren't necessary for hobby stuff. However, Fluke meters are top o' the line and built to a higher quality standard so it should last forever unless you run it over or use it improperly. Also, PS@ doesn't need a modchip anymore. I don't know about the Xbox.

Are 0.7mm ffc ribbon cables used by some displays some kind of jewish trick to make people pay more money? It is not standard and you can't find sockets for them

>>2757751
I think, it determines the voltage on v-out to be exactly of that amount to cause voltage drop of 0.6v across R2. Even in constant voltage application, it's kind of working as constant current

>>2759465
Is amplitude control a requirement? Or can I just use shorter pulse widths in place of lower voltages?

>>2759561
It's good to have that control as everyone's skin and pain threshold varies. Gives more granular control to fine-tune the effect.

>>2759494
If you want something cheap, and takes a long se you can get this:
ANENG A3005a
https://www.aliexpress.com/item/1005005492246379.html
Is it any good? for electronics, it's really bad (milli volts, or measure current, and the ranges are pretty bad), it is a gimmick, but it should do ok for continuity and it's easy to see the voltage when the display is on the probe.
Lost of meters are good at various things, like if you were interested in batteries or solar, you should get a DC current clamp (however most affordable clamps are AC only, and DC is a premium clamp, but I have seen $100+ DC clamps on facebook marketplace and local auctions for $30, typically missing probes).
There are also scope meters like this one (and there is a cheaper one he made a previous video, and you could buy an expensive 2channel 200Mhz scope (competes with benchtop scopes, and it's isolated from AC so less chance of blowing up your scope from accidentally measuring mains, I think?) + signal generator for around $200, like the this (and it works like a multimeter):
https://www.youtube.com/watch?v=-A21ORLuVVc
However the 9999 count meters are all fine, and every single 9999 count meter you can find for $30~ are typically using the exact same chip, with features like temp (I like temp because it's useful for testing your soldering iron heat), sine, batteries included, size, and etc being different.
I bought a AN870. It's not bad, but it's not premium either (I broke a piece of plastic in the battery case, it still works fine, I attempted to return using aliexpress but they wanted me to return using a non local most office that's an hour drive away, too much work).
Fluke is a good brand, but some offer rubbish specs because they are designed for electrician, HVAC and Automotive jobs (you can trust the meter to not electrocute you measuring higher voltages), the electronic engineering scopes (Brymen/UNI-T) have better ranges (similar to the 9999 meters) but are expensive.

>>2759571
>There are also scope meters like this one
I forgot to include the video:
https://www.youtube.com/watch?v=kpzNVsSmUu0

>>2759481
The relays will burn out.
They are also not fast enough.
Motor will be fine though, as long as there is no semiconductors in there.

>>2759494
Get a bench meter instead of meters intended to be used at the top of telephone poles by field service emergency personnel. I don’t know how this one-handed cat IV field meter meme got started, but I’m sick and tired of it.
Anyone pushing handheld meters to benchtop or lab work is a scam-artist or a know-nothing larper.
Picrel. Also, I hate philips.

>>2759580
Bench meters are fine if you're glued to a bench.

>>2759571
Oh yea and I forgot to mention analog discovery boards, they can function as very weak scopes, but the main interesting thing about them is that they are very powerful digital analyzers, you can connect like 10 probes or something and anaylze digital circuits, so like UART or usb stuff (but you can buy a $10 digital analyzer that works for usb 2.0, and uart, the Analog discovery is useful because the resolution is high enough that you can see degradation in the signal or something like that, and you can use the analog discovery as a multimeter (however it is connected to your PC, which could be an inconvenience, and I don't remember if the analog discovery is isolated from mains, or you need a thing for that, I am not a expert in electronics at all, never made a circuit).
Overall just check out the used market / local police auctions or liquidation auctions, you might find something, and you should learn to understand the specs of meters yourself by watching reviews and how to use a meter.

>>2759583
The last time I used mine, I had to work under a raised floor where they kept the mainframes.
The guy is working on xboxes. I assume he’s not working on them in-situ at people’s houses while they’re playing games on the bottom of the dusty entertainment center.

>>2759571
>and takes a long se
I meant to say "takes a long time to trigger continuity", if you bought a 9999 count meter, the continuity is better but isn't going to be good enough to drag across pins quickly (I think the pen is 1 sample per second or less, while the aneng is 3 times per second), you need to slowly one by one test the pins, but there are no cheap continuity testers with super fast sample rates (unless you DIY).

>>2759590
That's true, Anon, but both benchtop and handheld DMMs should have a place in your toolkit, and if you're working with mains then a clamp meter is good to have as well. Since Anon said he's just starting out, a handheld meter will give him basic functions and mobility at a low cost.

>>2759595
I’m going to be real with you. I hardly ever work with mains… and when I do, I have

• an incandescent light bulb
• a neon indicator
• a outlet tester
• a kill-o-watt meter clone that tells me the exact voltage, frequency, wattage/current, etc.

These are pretty much designed for mains activity, bench or otherwise. Highly portable. I’m not clamping on a meter at any point. These are for testing shit like individual wye legs by industrial electricians. Also, if I need to, bench meters can also test mains stuff when the PS2 or Xbone is open, sitting on the bench.

>>2759592
This continuity test speed/latching nonsense/meme was invented by dave jones. It’s the equivalent of wine-tasting exhibitions. I’ve never has a problem, and you’re right, you can build an ultra fast one in minutes if you ever need. Or use a sample and hold, or pull out your logic probe, etc.
It’s pretty much a non-issue.

Am I understanding photoresist pcb etching correctly. After UV process, you apply "photoresist developer" (Sodium hydroxide) which removes the exposed part of the photoresist film and leaves bare copper right? And after that you use sodium persulphate to etch away the copper?

I have an inkjet printer so this seems to be best method for me. I am planning to use transparent papers (that it says it works with inkjet) that are used for printing stickers. I imagine this should work?

>>2759580
Here’s another issue. I hate LCDs and I work on a dimly lit bench instead of an office designed with bright lighting standards of 1967 for pushing paprts all day long.

I can’t afford an oled meter, but picrel is the meter I would buy. It uses LCD with a back light and big digits so I can read the fucking thing. Less mistakes.
Zoomers should love this shit too. it’s “darkmode” compatible (now that The CRT look is back in fashon) instead of the stupid apple lisa white-paper look. It’s even got bluetooth audio so you can listen to your nightcore and lofi. Everything a zoomer and a boomer needs.

>>2759602
> pcb making
Yeah, that sounds about right.
I’m not sure an inkjet printer is dark enough, we used lexmark (ibm) laser printers in the past, which laid down ridiculously thick toner layers.
We used UV-transparent clear film, and “glass” plates that we sandwiched the pcb and film between to expose it.

>>2759608
yeah inkjets seem to be ideal but I don't want to buy a second printer just for this, and they cost a lot

Maybe I can use the library printer if mine doesn't work good. Those huge ones are always laser right

>>2759561
>Is amplitude control a requirement?

it's an absolute necessity.
every time you connect someone, you must start at zero and ramp it up slowly.
just like at the dentist, instruct them to "scream loudly if it gets too painful"
to masochists, say "scream loudly if it's not painful enough"

>>2759601
>It’s pretty much a non-issue.

nonsense.
say you wanna find what pin a test point is connected to.
if you have a bunch of 80-pin chips you might spend 2 minutes on each one carefully touching each pin and waiting half a sec for a beep.
but if you have an instantaneous result, like using pic-related, you can quickly drag and find the exact spot in seconds.
frustration level decreases 100 fold.

>>2759580
I portable one thats 'decent' is $20-50. The cheapest bench one i can find online is over $100. I know thats gonna mostly be shipping and i should locally, but everywhere i look costs even more for bench ones since they're niche. Craigslist does not have any listings for one either.

can i connect crystal oscillators in series or in parallel to combine frequencies?

>>2759695

not without electronics to mix them and create beat frequencies.

>>2759494
I am a fan of ANENGs but I also added this to my collection:
https://www.amazon.com/Ranging-Digital-Multimeter-Battery-Alligator/dp/B07R47RYPW

It is an absolute beast! 20k count and the quality is top notch! It is big and heavy and extremely sturdy with a nice kickstand (unlike the flimsy AN8009), high quality plastic, and yes it is 20,000 count! Super accurate, beats most bench meters. And it is cheap.

>>2759648
I have a $5 harbor fright multimeter and it’s continuity test is as as good, if not better than my fluke.
I used to test huge wire-wrap boards according to the connection list, so I do know something about it (even if that is that one should avoid drawing that short straw).
Probably 90% of he multimeters out there use the same cloned fluke chipsets, I’ve never run into onethat has any problems, and I’ve seen a lot.

>>2759679
Take your time. I’ve been doing this for decades, and if you’re going to be doing this for a while, at least invest a bit into your most important tool. It’s your first and primary diagnosis tool.
It’s a little bit like writing code… 80% of the time is spent tracking down bugs. In this case, faults.

Although I’m already retired, I really want that benchtop Aneng benchtop now that I found it.
It even has a capacitance test.
I can listen to my carly simon and neil diamond compilations on it. What’s better than that?

>>2759726
>I can listen to my carly simon and neil diamond compilations on it. What’s better than that?
Thats what makes me think its a POS.
Chinese branded gadgets that throw weird extra features on those kind of products almost always do so to get more sales from naive people.

/ohm/, tell me about your duh moments. Stuff you should have or could have known but somehow you didn't. For example, after so many years I finally realized that a TH LED has a flat side which is the cathode. I only knew that the anode lead was longer and when I cut the leads short I still kept the anode longer so I could tell which is which.

>>2759748
> POS
It’s probably got the same board and chipset as the handheld. I’m sure the bluetooth audio is a separate thing they just threw in there.

Now, what would be even better? If the handheld meter had a pcb edge/docking connector on the back, like a video game cartridge. Then you just buy the handheld, and to turn it into a benchtop by buying a $20 optional benchtop case. Then the handheld slides in, you close the hatch, and all command functions are transferred to the new docking station. Maybe you could even get a docking station with VGA out so you could really see it well.

Sadly, this only works if they can’t sell you on both devices, and it assumes that it would actually be cheaper and the whole meter doesn’t cost them $2.00 to make in either case.

Oscilloscopes have the same problem. 80% of the cost is an unwanted, under-powered tablet computer that something like an iPad 1 trounces in graphics and compute capability.

>>2759639
So you can't just drop the duty-cycle to 0.1% and have it be barely detectable?

>>2759866
>drop the duty-cycle to 0.1%

try it and see.
what i found is that if you try to really shorten the pulses, the transistors will fight you.
they really wanna spill their load, so they'll stay on longer than what you tell 'em to.

>>2759877
Was only going to go for 10Hz or 50Hz, since I've heard those are the important frequencies to hit, which would give me a lot of headroom. Still haven't found an IC that can independently tri-state that many pins at such a high voltage though. Guess I'll build a discrete version with some totem poles for testing.

>>2759602
A few things:
Firstly, transparencies are a pain. If you're working with dry film photoresist, it is often of dubious quality from chinese vendors, with thin spots and such. I found I couldn't get enough contrast with either inkjet transparencies nor with an MSLA resin printer to develop it with sufficiently high contrast to not get fuzzy edges at 0.5mm resolutions. I think laser toner is generally better than inkjet on a transparency, but I'm not certain. If you're using dry film, you want sodium carbonate (washing soda), not sodium hydroxide. I think hydroxide is used for presensitised boards, but those are expensive and kinda hard to get. Watch Big Clive's videos on the topic for a good overview on etching and exposing with a transparency. Liquid photoresist is nasty stuff, avoid unless you have a fume hood with a spin-coater in it. You may find the toner transfer method to be easier and more reliable, but I've heard it either way.

Secondly, your etchant of choice isn't that ideal. Persulfate etches somewhat slower than ferric chloride, and the longer your board sits there the softer the etch resist becomes. I find that ferric chloride can etch a board in 5-10 minutes, with mild heating and agitation (from an ender 3's heated bed moving back and forth). I made that ferric chloride by dissolving nails in hardware store muriatic acid, it's quicker if you bubble air through it. I can also regenerate the etchant by electrolysing it extremely slowly with a platinum anode, then cleaning off the copper dendrites that grow on the copper cathode.

Finally, if you have any CNC device other than a printer, you're better off using that. I use a 3D printer with a laser bolted to it, in order to burn away an etch resisting spray-lacquer. But a dedicated laser engraver, CNC router, or other flat XY machine you can bolt a laser onto would work fine.

>>2759481
Won't work. You need 6 switching elements to make a 3-phase inverter, and you need appropriate shoot-through protection. Better off just modifying an ESC with high-voltage transistors and faster gate drivers, plus whatever circuitry you need to isolate the high voltage from the rest of the circuitry. Or just buy a chinky or 2nd-hand VFD. Or a rotary phase converter, lmao.

>>2759607
If you're buying a bench meter, you might as well get one that's also a 4-wire LCR meter. Because somehow they don't make kelvin sensing handheld meters.

One thing I realized after a modest time of messing with PCB is adding parts one by one or in small groups is easier to deal with compared to wiring up the entire schematic and then dropping the entire clusterfuck of netrats on the PCB and spend hours untangling them. For example I tried adding an MCU and a shit load of bypass capacitors first, routing them, then do another chip, and then make all the connections between the two chips, etc. Also hiding the ground nets helps a lot since they are all connected to the bottom ground plane via vias anyway and that reduces clutter. So I hide them, route everything else, then unhide them and slap vias on them.

>>2759580
Bench meters are cool but take way too much space. And a lot of times you need 2 or 3 of them at the same time. Would you put your bench meter at the back of your desk and use really long leads or would you move it around all the time? That's the beauty of MMs. You can spread them and move them around your desk/bench as needed.

>>2759889
>2nd-hand VFD
yeah, looking at the used markets prices aren't as bad as I thought.
If I want to drive a 4kw motor at higher than 50hz line speed, do I need a bigger than 4kw VFD or does it just trade off torque for speed and stays at 4kw power?

Anyone own these probes? I'm (>>2759494) looking to add these to this an8009. Reviews look mixed, but it's hard to know if they're legit since theyre all in broken English. they're the MaAnt p22a. I picked them cause of the silicone wire and good coated tips, plus they don't have crazy bulky plugs like a lot of other "premium" leads. To be clear, I'm cheap af and still only paying $12 for them. I know generally premium ones by fluke or whatever would be $60+, so I'm not under any delusion these will be as good as those... Just better than the included ones with the an8009.

>>2760161
i have no idea but if you want to spend a little extra money, buy genuine ProbeMaster probes. they're only $25 from their website. i have a pair and they are definitely worth the price.

>>2756613
Besides w2aew and VK3Y, there's ImsaiGuy. He explains everything in a really simple way for beginners and experts alike, almost no math involved and he's a HAM radio guy too. I also recommend downloading the Art of Electronics 3rd Edition and Art of Electronics the X Chapters pdfs. They're great and I also bought the books irl.

>>2760023
Well, I like the big digits, I do lot of measurements, so it’s the most important thing—visibility.
You don’t hear gamers complaining that eldon ring didn’t come out for the gameboy.
Also, I’m the kind of guy that would never be able to find the damn things. I know one handheld is in my electrical toolbox. Others??? I have 6 tape measures and 6 olfa knives and I can only ever find one of them after an hour of searching. I’m sure some are sealed up in walls, like my other handhelds are sealed up in equipment panels somewhere.
That darkmode of the aneng display is downright cool.
Now lets talk about something important.
What are you going to show chicks when you bring them over to see your bench? You gonna pullout some LCD handheld with single feeble blue/white led off to one side? No, you’re gonna want to flip on the big’un so you’ll look like a nuclear phy$ici$t and play some barry white over the bluetooth speakers.

>>2757109
Yes, unless you're working with something in the RF domain or you got some metrology stuff. Otherwise it's not going to affect you in any way.

>>2757127
>they'll move me to soldering
You'll get really good at soldering up until a point, from which you won't improve much over 5 or 10 years. You will hopefully find another position within the company or get hired somewhere else as engineer or tehnician

>>2760200
>"Ladies you are looking good tonight” Then just walk away.
I like this new /ohm/ character.

>>2760200
is there /ohm/ themed porn on pornhub?
>do it on a bench full of /ohm/ approved DIY projects like this one: >>2759098
>babe wanna see my cockroft?

>>2760233
If I see a properly set up oscilloscope I'd pay for it just for the lols

>>2760269
I can see some sort of “Contact” plot with alien radio signals and sophisticated radio monitoring equipment, only for morph suit ayys to bust down the door and start a group sesh, making animal-crossing style bleep bloop noises all the while.

Is heat the only limiting factor to getting more power out of an electric motor? Like with a good enough cooling system, would it operate just fine with 2-3X the amperage?

>>2760384
Similarly, things like transformers have a “duty cycle” when being used for, say, welding.
1 minute on, 5 minutes off to cool.

>>2760384
>Is heat the only limiting factor to getting more power out of an electric motor?

It's far from the only factor, but it's overwhelmingly the most significant one. It's very easy to run into thermal limitations, either hard ones, like the windings immediately burning up, or soft ones, like reduced lifetime from running 15°C hotter than usual.

Once you're putting enough amperage through a motor that you're running into heat problems, the easiest way to increase power output is to spin the thing faster. Power is force over distance (in this case, torque and rate of rotation), so getting the RPM up will get you a more powerful motor. You then start to run into mechanical limitations, again, either hard ones (the rotor grenades itself by throwing the windings off) or soft ones (reduced bearing life). In specific cases, you may also get into trouble with the increased voltage required to increase RPM, particularly with brushed motors.

Generally speaking, though, yes, you could do something like put a water jacket on an induction motor or inrunner motor (the latter is actually seen somewhat commonly in RC boats) and greatly increase its continuous operating current. You'd still need to worry about heat in the rotor. It's significantly harder to get heat out of that.

>Similarly, things like transformers have a “duty cycle” when being used for, say, welding.

With the right design (like the oil-bath ones seen on power pole transformers), transformers can generally be run right up to saturation without problems. There's not really anything you can do at that point, since the operating voltage and frequency is usually dictated by the requirements of the rest of the system.

>>2760384
In addition to what the other guys say, I'll add two points:
1: the permanent magnets in the rotor of a BLDC/PMSM/stepper have a curie point, which may well become the limiting factor if you water cool your motor for more amps, or even just run it with a few more amps with decent airflow for 30 minutes.
2: magnetic saturation can be an issue with motors as well as for transformers. I haven't tested this, but I suspect with an RC prop motor running stalled could well enter saturation.

Furthermore, brushless motors can generally be driven from an arbitrarily high source voltage, so long as your ESC can PWM the coils fast enough to limit the current. This will allow you to reach very high speeds, which with a gear reduction will generally be more efficient for extracting power compared to running it slow, where ohmic losses are more significant. Though you could get the same efficiency increase by just rewinding the motor with more turns of thinner wire. Either way, you don't really want to be doing significant mechanical work while your PWM duty-cycle is particularly small, as the peak current through the FETs will mean the ohmic losses in your driver will be higher compared to with a motor that can spread that current across a longer on-time. Because power goes by the square of the current. I'd only overvolt your motor with low-duty-cycle current limiting PWM to such an extent if it was just for the purpose of getting your motor up to its faster working speed, at which point the back emf would take up a lot of that higher voltage and allow you to PWM at a higher duty-cycle.

I want to make (not buy) a BLDC motor with 100 Nm of torque at low RPM (up to 1000). How to design one? I can buy stators of different sizes on aliexpress, same with magnets. I can machine a rotor from aluminium. What size of rotor I need, what magnets I need, what windings I need to get a motor with requested parameters? Don't write if you don't have answers to my specific questions, I'm tired of retarded electrical engineers who don't know how to design a fucking motor. Imagine me as a civil engineer not knowing how to design a fucking rc beam.

>>2760478
Sounds like you want to calculate the torque from a given stator and rotor design. This involves calculating the magnetic fields and doing some funky vector calc. It might be doable to get a torque value from a series of approximations about the magnet's strength and the magnetic field density you'd get from x amp-turns going through y area, but it could easily be out by a factor of 2. The closest thing I ever did was failing a paper on EM that introduced me to the magnetic dipole vector that could be used to calculate a torque, but in reality calculating the dipole strength isn't that easy and you still need the magnetic field strength for the other part in order to calculate torque anyhow. So I think you're best off using some sort of CAE magnetics simulation software. IIRC there's a FOSS one.

Also while chinky rotors might be ok (assuming you don't want a fancy sort of silicon steel), don't trust chinky N52s to actually be N52s. FYI you'd be better off with an iron rotor, since it will close the flux loop of the magnets. If you use a halbach array it doesn't matter as much. Aluminium should be fine for the side-walls (assuming it's an outrunner), and for the bracket that holds the stator.

t. trying to make a hub wheel motor from a 6215 agri-drone bldc

>>2760384
>amperage
current

Noob here. Trying to see if I can repair an old Western Digital HDD that stopped spinning years ago. I've watched a couple of youtube videos that mention the problem might be cause by a malfuncioning diode (zener diode). When test the diode using my multimeter in diode mode I get: 0.6 (forward), and 1.6 (backward); I really don't know what to make out of it. Picrel is the board. Also, is it me or the screw holes an J1 pads look all smoked?

These are the videos I found with the same model and problem (not spinning, no noise)
https://www.youtube.com/watch?v=qdXW-DUSsh8
https://www.youtube.com/watch?v=2YZ0LBhFCZY

I'm tryna fix my parents kitchen fan hood controller but I can't find the name of this component, wtf does L1 _S_ sideways ON/NO mean bro?

>>2760611
>I get: 0.6 (forward), and 1.6 (backward)
Seems bad. You should also check U9,C66,C70 (bottom right near power connector) for stable voltage output.
>is it me or the screw holes an J1 pads look all smoked?
It's tarnished from contact with the air. Use a pencil eraser to get rid of it.

>>2760612
ON Semiconductor is a chip manufacturer. Could be that, but usually their logo is set in a full circle.
>>2760615
It looks to me as if the plating is gone. Not uncommon with prolonged contact with a dissimilar metal washer.

>>2760619
>It looks to me as if the plating is gone.
It's definitely corroded, but I was replying the the HDD anon.

>>2760615
>Seems bad

disagree. looks normal.
you can verify that it's normal if you read 12V across it when you plug it in.
but clean the J1 and J6 contacts before.
J1 is the read/write head plus arm control, and J6 powers the motor.

>>2760627
Isn't J6 gold-plated?

>>2760627
>clean the J1 and J6 contacts

forgot: also clean the mating connectors.

>>2760633
what are those?

I've been trying to get an old and bought-as-broken printer to work. It's an NEC Pinwriter P2 Plus, in some markets known as P2200. The PSU is broken beyond my capabilities to repair, as I've tried for three times now. Last time it also blew two caps on the logic board and I pray to God it didn't burn out any ICs since those are close to impossible to source.
Finding a replacement PSU seems nigh impossible, too. Since I really want it to work again, I've settled on jury rigging. The original gives out two rails of 5V and 34V, as well as each having their own ground rail.

Right now I'm trying to figure out how much current the 34V rail will draw. Since it has to power small motors for carriage and feed and the needles of the printhead itself, which are probably some form of solenoid, I guess I won't get far with the chinkshit board I just ordered from aliexpress, which is rated at 35W, so roughly 1A.
The printer manual doesn't state the draw per rail, it only states a combined 0.6A on 230V AC. That could, in the worst case, mean around 4A on the 34V rail, since the 5V are probably solely for the logic, which won't draw a lot.

34V is rather uncommon for power supplies, so I just asked myself if I could just use one with 36V. Any idea whether that'll have any unforeseen consequences, such as turning the printer into an puddle of molten lava?

>>2760634
the connectors that mate.

>>2760652
so they can make baby connectors?

>>2760636
>The PSU is broken beyond my capabilities to repair, as I've tried for three times now. Last time it also blew two caps on the logic board
Sounds like a short. Find it before you try to power it up again because the same thing will happen with a new PSU.

>>2760693
I really hope it's that. The caps don't seem like they've been replaced in the past, so they're likely as old as the printer itself. I've ordered replacements (22uF, 63V) in hopes that they were just bad themselves. All three times I tried to fix the PSU board it failed the same way, namely a big cap blowing up (10000uF, 63V).

>>2760636
Around 34 V is as common as dirt for power supplies in printers, I think all those cheap-ass HPs are already 32 V. You can get adjustable, and just tweak their V with a trimpot.

You probably want to go up in Amps, not down, if you’re replacing it.

When testing, you can limit the current with an incandescent light bulb in series with The mains L. I even have these mini night light bulbs that are only 7 W which work great for low powered stuff.

Sounds like that 10,000 μF cap was getting AC probably. Anyway, when I replace a cap on something important I always go up in voltage to the biggest cap that will fit. And go with Panasonic from a reputable source.

What’s so great about this printer anyway?

>>2760636
>Last time it also blew two caps on the logic board
Were they tantalum?

>34V is rather uncommon for power supplies, so I just asked myself if I could just use one with 36V.
Just get a 36V one with an adjustment pot

Those are the specs for a model of the same series https://archive.org/details/manual_P60_SM_NEC_EN
2A at 5V sounds right for the amount of logic on the PCB of the P2
Motors shouldn't draw too much but the hammers will do it in bursts so it needs good peak power

The best idea would be to test the power supply outside of the printer with some power resistor/lamps as loads

If I'm doing a TDR on a twisted pair cable, do I cal my VNA for the 50Ω from the balun back to the instrument, or for the 150Ω (75Ω per wire) nominal impedance of the cable under test?
My VNA came with a 50Ω SOLT kit, but using that gives me pretty shitty results in this instance. Instead, I "made" my own SOLT cal kit with a homemade LCR filter for the Load cal, and a 1M piece of the cable on the secondaries of the baluns for the Through.

>>2760730
>power supplies in printers
Didn't know that, only looked on ae. Got a cheap 36V/7A in my cart but wasn't sure.
>biggest cap that'll fit
The last one was 100V and it looks like it survived, but I unplugged it after the loud bangs that I now know came from the smaller caps.
>What’s so great about this printer anyway?
Nothing in particular, I just like it. I mean, it's a great model as far as dot matrix are concerned, it's got 24 needles and can do decent grayscale graphics from what I've seen on youtube, but mostly it's just my affinity for older tech.
Just to name one example, the logic board is absolutely beautiful. It's got a mix of thru-hole and smd components and the traces are in right angles and not curved, as they often are on modern pcbs.
Power board: https://i.imgur.com/dQPgtvO.jpeg
Logic board: https://i.imgur.com/uYmPH4k.jpeg

>>2760733
>Were they tantalum?
Not sure. Looks like normal electrolyte to me, but they have a metal housing. Pic related, the grime is some sort of fibers from inside the caps.
>Just get a 36V one with an adjustment pot
Would be the next step if the cheap one doesn't work

Thanks for your thoughts, will do more research tomorrow. It's getting late on this side of the pond.

>>2755510
i need to measure single phase 60 hz AC mains voltage and current, 120V RMS and 20A. what's the most cost and space effective way i can do this? voltage and current transformers are an option... but are bulky and i'd have to dick around with phase delay nonsense.

>>2760757
>voltage and current transformers are an option... but are bulky
They can be pretty small tough
>dick around with phase delay nonsense
You don't need to unless you want to measure power, for which there are ready-made ICs and complete circuits

>>2759098
did you drill the glass

>>2760757
touch with your tongue and use mehdi's pain chart to figure out current

>>2760797
There's a sheet of acrylic on top of the glass. You can kinda see it near the red alligator clip.

>>2760636
I’d try it with a 36V brick and 3 series silicon diodes.

>>2760757
Pull the guts out of a killawatt meter.

How do I fix Exit Status 1 for Arduino IDE. Google doesn't help and everything I try doesn't work. About to ditch Arduino for plcs or something else if this is the shit I have to deal with as a beginner.

>>2760843
You probably want >>>/diy/mcg

>>2760843
>>2760885
Also, it’s arduino. It’s not called “tarduino” in /mcg/ for nothing.

If I put a cap in series with a solar panel with telescopic optics, then attach it to the input of an audio amplifier, can I listen to the sun? Can I hear a fart on Uranus?

>>2760923
Yes.
We use them to determine the switching frequency and ripple in those LED light bulbs. Definitely fast enough for audio.

>>2760925
Could I make a 300bps BBS with a tethered LED balloon over my house? Will I need to implement Zmodem protocol for warez transfers?

>>2760923
>can I listen to the sun?

why limit yourself to our boring solar system?
build a gravity-wave detector and listen to stars exploding, black holes being born, your fat mom falling down across town.

>>2760636
>34V is rather uncommon for power supplies

like the other dude has said, many HP bubble-jets use external +16V and +32V power bricks at around 900mA each.
often see them at the thrift store for a couple of bucks.
noticed your printer seems to have 2 grounds, SG and DG. so you may need to separate the grounds on +34 and +5. check against the original.

in any case, i suspect the caps died coz of over-voltage, which also killed some other parts. so you may be shit out of luck.

does anybody sell wirelessly controlled power switches that can handle high-ish currents? (somewhere around 1A)
i need to remotely connect a battery to a video transmitter for an RC project

>>2760934
Geez, use 2400 baud and v.42bis, so you can worry less about things like zmodem.
At least you didn’t say kermit.

>>2760923
You'd be better off with a conventional photodiode in photoconductive mode. A clear non-white LED would work in a pinch. You could use two different colours of LED (different bandgaps of photodiode) to pick up different wavelength ranges.

>>2760941
Can't you tap off a servo channel from the RC receiver and lead it to a MOSFET? /rcg/ may no more about customising an existing RC controller system.

>>2760938
Separate ground wouldn't be a problem, as I'll be using separate power supplies for 34V and 5V. Already ordered a cheap USB wall wart that I'll just break open and hardwire to 5V.
I really hope it didn't fry anything. At least none of the chips are visibly damaged, since parts of the metal housings of the caps flew over the board though, it may have shorted something though. I'll clean it up to the best of my abilities before plugging it in again. Fingers crossed.

>>2760756
To follow up on why I like such old electronics, here's a curious section of the logic board.
All three ICs are six channel logic inverters with almost identical parameters, made by the same company (Fairchild Semiconductors).
The thru-hole has open collectors and the datasheet explicitly warns that they need pull-ups for correctly defined high levels, and who would've thought, they end up in an array of resistors. Since that seems to be the only big difference between the thru-hole and SMD, I guess the SMD was more expensive and was used for its smaller footprint, whereas the thru-hole had enough space and it was cheaper to add a few resistors than another SMD.
Thru-hole: https://www.alldatasheet.com/datasheet-pdf/pdf/50889/FAIRCHILD/DM7405N.html
SMD: https://www.alldatasheet.com/datasheet-pdf/pdf/51019/FAIRCHILD/DM74LS04M.html

>>2755510
does anyone have experience with / can tell me whats the difference with electronics books labeled as "indian edition" as opposed to the "non indian edition" ????
i have ordered "Fundamentals of Analog Circuits " by Floyd / Buchla and somehow missed that there is a small flag saying "This edition is manufactured in India and is authorized for sale only in India Bangladesh Bhutan Pakistan Nepal Sri lanka and the Maledives" the book is from 2012

>>2761072
>Fundamentals of Analog Circuits: North Sentinel Island Edition
The guy who put LEDs in shoes back in the day got the Indian edition too.

How do I tell, what frequency to drive a high voltage transformer at?

>>2761122
https://www.youtube.com/watch?v=tjbK4LsOQRk

>>2761072
I think the Asia-Pacific editions are cheaper paperback editions compared the standard editions, I'd guess Indian editions will be similar to that. Not that cost should matter, get them from libgen.

>>2760756
Those SIPs. Ohhh… what’s the part number on those? Imagine a world where they took advantage of the most efficient and best way to package something.
Like a single op-amp in a 5-pin SIP.
Or a fixed 10× amplifier, single buffer in a 4-pin SIP.
Or multi-emitter transistors.

>>2761036
That pcb is a good example of terrible design with those squared-off traces. It looks clunky, and would be terrible for carrying high frequency signals or spiky signals (as this thing does).

Also that is National Semiconductor.

>>2761195
It's just a limitation of the EE-CAD tech that was available at the time. I guess it worked well enough and was cheaper than routing by hand. Bet it could handle 10MHz signals without a problem, probably even 100MHz.

Pretty stupid question here, but I bought a cheap chinamang kit to practice soldering, but the instructions only came in moonspeak.

Are the pads on the back of the transistors supposed to be in contact with the pads on the PCB (Q2-Q9)?
I also got some extra screws which really doesnt go anywhere except these holes. Am I supposed to screw them or the MOSFET down? Only got 4 screws between 8 transistors and 1 mosfet.

>>2761198
I remember, it was terrible. Part of the problem is they radiate and you pick up the signal on other traces. We’d wait all night for the autoroute to complete. Then, when it didn’t work, you’d re-roll. When you complain, you were told that pcb routing was an “entry level position” and that we’d all get replaced by automation some day.
Just like today, replacing “automation” with “AI” — scams and business trends are definitely cyclic.

>>2761209
What a waste of resources.
Anyway, no.
You’re supposed to bolt everything with a hole to a heatsink. That’s why hole.
The other thing you see is free-standing with maybe some silastic if you’ve over-rated your transistor. Zheng always uses “Absolute Maximum Ratings” if not under-rates everything.

>>2761209
yes, you screw them to the PCB. You don't need a heatsink unless they get noticeably warm (debateable)
>Am I supposed to screw them or the MOSFET down?
yes
>Only got 4 screws between 8 transistors and 1 mosfet.
chink'd. hit up your local home depot

>>2761195
>>2761198
It's a matrix printer with a parallel interface from the late 1980s, not a ghz-class digital spectrum analyzer.
If it didn't work, it wouldn't have sold.

>>2761072
What >>2761165 said. Some publishers have low cost editions of their books for second and third world countries. I've got a copy of The Art Of Electronics from Cambridge Low Price Editions that formerly belonged to some tech school in Bangalore, India.

>>2761190
https://www.alldatasheet.com/datasheet-pdf/pdf/6536/NEC/UPA1437H.html
Appears to be a Darlington

>>2761292
> parallel interface
Yes, that was a slow interface because of the interference from the parallel port wires with themselves, and shitily designed pcbs. This is why we use tend to use differential pairs nowadays, the IBM parallel port was one of the biggest drivers of that.
> drivers of that. Get it?

>>2761296
Power SIP porn.

I want to build a star finder that I can fit on my telescope.

Thinking of a two stage project
Phase 1. Use it to display basic azimuth and altitude of stars. Zero on Polaris, and then it tracks relative movement from there.
Phase 2. Use it to turn stepper motors to counter Earth's spin

While gyroscopic drift isn't a big concern for Phase 1 (I can always get it to point in a general direction and then zoom in), it's a big problem for Phase 2. Shy of strapping a TPU and using machine vision to control the steppers, are there any better gyroscopes on the market?

> currently what I'm looking at
Pololu AltIMU-10 v5 Gyro, Accelerometer, Compass, and Altimeter (LSM6DS33, LIS3MDL, and LPS25H Carrier)
https://www.pololu.com/product/2739

>>2761390
BMI160 is cheaper but has comparable specifications to LSM6DS33. Accelerometers point down without drift. I suspect that two accelerometers can give you the azimuth& altitude of an equatorial mount.

>>2761390
How hard is computer vision really? I hear there's openCV for ESP32s that's used for facial detection.

>>2761339
I mean, it's not like the needles were capable of any speed record to begin with. I don't think the interface was the bottleneck, it's been rather held back by physical limitations of the solenoids moving the needles. other technologies, such as laser and inkjet, had to accelerate and move less mass swhile printing, which obviously made them a lot faster.
Matrix printers have their advantages in other fields, such as making multiple carbon copies in one go, for which they are still used by many doctors and banks.
Also, it's easier to refill the ink by just opening the ribbon cassette and drenching the ribbon in a suitable ink. Inkjet and laser cartridges are so tightly sealed that most of them are almost impossible to open without damage. If they're not specifically intended to be refilled, you can basically throw the printer out once the cartridges are out of production.
By the way, I'm pretty sure the Darlington SIPs lead to the ribbon connectors for the printhead (the two horizontal ones in the top left of the logic board pic in >>2760756). I think they power the needle solenoids.

Also, the replacement caps have arrived. Gonna get a shower, then clean up the board and install them.

>>2761461
>openCV
Can be a bitch to install without falling into dependency he'll, but its pretty straight forward once it's installed. Even training new models can be easy, if time consuming.

>>2761443
I'm just worried if I get to a stage where I'm taking long exposures with a DSLR it's going to cause blur. I guess exposures probably wouldn't be longer than a minute anyway, when drift occurs over an hour plus

>>2761519
Template matching in opencv isn't quite right. The displacement between successive frames will be small, so there's no need to make an intermediate image 4 times as many pixels.

>>2760730
>>2760733
>>2760838
>>2760938
Just ordered one of those power supplies you'd get with cheap 3D printers for 25 bucks incl. shipping. It's 36V and rated for 100W, so close to 3A.
Those power supplies have a small pot next to the screw terminals with which you can adjust the voltage, from what I've read they go roughly 10% above and below, so it should do the trick.
The 150W version was only 2 bucks extra but I'm a bit restricted in terms of real estate inside the chassis, since I want to install it in place of the original one and not have an external power brick dangling around.

>>2761390
>>2761443
If that's of any concern, the BMI uses less power (925uA in full operation, 180uA if accel on/gyro suspended) and it has a few power saving modes down to a few uA if sampled at a low frequency and low amount of samples averaged.
At 8 averaged samples and 6hz sampling rate, you can get as low as 10uA.
As a side note, this low power draw is why it was used in the Pokewalker/ WiiFit pedometer as step counter.

>>2761514
> high speed needles not that fast.
The concern is that even those SIP transistors can run at 1 MHz by turning on and off suddenly generating high-frequency noise.
The parallel port drivers are the same, making the length of your parallel port cable very limited before it basically stops working because all the lines interfere with each other. Hard right angles on the PCB are not your friends.
> all the lines
Transmission lines, that is.

Even a household light switch generates enough high frequency suff to kill your LED’s light bulb electronics, not including switch “bounce” noise, which repeats the whole high-frequency switch action and arcing.

>>2761819
https://resources.altium.com/p/pcb-routing-angle-myths-45-degree-angle-versus-90-degree-angle
sorry, but it seems you've been deboonked

>>2761914
> altium has limitations, here’s how we justify them with an electrical engineer that also maybe created the batterizer and solar freakin roadways.
Not only are right angles unnecessary and wrong, but so are sharp 45. It’s because maybe they managed to become electrical engineers, they never learned that the shortest distance between two points is a straight line. Look at that shit in picrel. LOOK AT IT!
A gameboy emulator on an 8K projector has less jaggies. Why? No good reason, other than incompetence and laziness.

Their argument is bullshit. They are implying that there is no effect up until 10 MHz but everybody knows it’s a continuum from bad: what they provide to optimal: fully curved and amorphous shaped traces. Where’s all this AI? I know where.

Ever built something on a breadboard from a PCB schematic and it doesn’t work right? Even an audio project? Those effects are found on a breadboard in spades. Yes, yes, in addition to other effects. But the breadboard actually has, in effect, acute angles. They should promulgate those as looking cool for sub-1MHz boards because it’s all goth/retro.

Even if my project is only going to 1 MHz, why can’t I use standard best practices that have been known about for decades? Well, I can, and I will. Feel free to follow some marketing guy’s excuse-o-blog.

>>2761914
I doubt they did any kind of test.
It certainly wasn’t with 34 V and high currents driving pure inductive loads like we have in this printer.
It’s altium that is debunked.

>>2761921
>best practice
There's a difference between making something work and going the extra mile.
Of course you could mill the frame from a solid block of tool steel, hand-pick the ICs, use aerospace-grade wiring and pot everything in epoxy.
Ever heard of the pareto principle? It's a conumer-grade product, designed for cheap and efficient mass production.

>>2761974
Yes, that’s the point. We’re just talking about something really easy to get right-like drawing a straight line at the “paper-pushing” stage. A child could do it. A curved line is not alien technology. Well, it didn’t used to be.
We’re not talking about 8oz, gold plated copper traces on 2 mm ceramic substrates.
Those printer PCBs are also cost ineffective to produce, they have to dissolve 95% of the copper, and then electrolyze all of it out of the etch bath.

>>2762095
PCB fabbing is machine time, it's not expensive. Designing the board takes human time, that's where the costs are. As I said above, those square traces come from EE-CAD software, which takes an order of mangnitude less time to route compared to applying trace and footprint tape to a glass plate like your pic. Add another order of magnitude or two if it's autorouted.

I don't think they even charge you for particularly large amounts of copper that need to be dissolved.

>>2762111
Watch dave jones exploration of modern autorouting. It’s not much better than it was in the 80s. Maybe worse.
You’re not supposed to just autoroute a whole board and ship it.
You just auto-route parts of it which may be repetitive, look at it, and fix it up.
Or fix glaring auto-placement errors, and let it auto-route from there. Or, manually place and route the critical parts, and let it autoroute ancillary support components.

Although I’m sure modern zoomers just directly ship 100% ChatGPT code for critical life-safety systems now. Good for population control.

Similarly, I see lots of seemingly 100% auto-routed boards, with vast unused areas and complete nonsense, like placing a single inductor on the other side of a board for a buck downconverter. It’s laughable. And in quantities of 1 million, that matters a lot. Especially when the stupid auto-router needs a 2 or 4 layer board for one trace out of a thousand on the .1 rev.

> charged for copper
Exacty… you’re not charged for it.
Get and receive your free copper! That’s *your* copper they are keeping.

PC Engine
1987
glorious Nippon acute angles folded over 9000 times

>>2762129
Imagine not keeping all your MELFs in the same orientation.

>>2762111
The problem is, modern zoomers don’t even really understand or know what a pcb is supposed to look like, so they see those lines in a sci fi movie and emulate it without any understanding of what’s going on.

>>2762139
Lesbian pairs 69'ing on the end.

>>2762124
>You just auto-route parts of it which may be repetitive, look at it, and fix it up
Yeah, but early boards couldn't even be manually routed with rounded corners, because that's a limit of the technology of that kind of ancient software. What, do you expect them to print it out on a transparency and draw in the curves by hand?
>like placing a single inductor on the other side of a board for a buck downconverter
If that means you can reduce the switching loop area, go for it bro. You may not like it, but that's what peak performance looks like.

>>2762139
pick-n-place is dumb, please understand

>>2762153
Yes. Meanwhile, us patricians are downloading Mitxela's trace melting script for KiCAD (which has no autorouting) and making beautiful PCBs:
https://mitxela.com/projects/melting_kicad

>>2762157
> technology limit
If I can get real-time life-like custom plumper porn rendered at 120 fps, I should be able to get rounded traces after a 24 hours of render time for 1 frame.
If they put as much effort into the tit bounce algorithm and semi-solid fluid dynamics optimizations, we’d be living in an electronics utopia.

>>2762163
>real time 120fps
>x random access memory
>y non volatile memory
ok now extrapolate those specs backwards by 40 years via moore's law

>>2762095
Your pic shows traces drawn by hand by an engineer. You're looking at multiple days of work at $50-$70 an hour. Autorouting takes little more than importing footprints, connecting them in the right order and clicking autoroute, which, as the other anon said, wasn't really there yet in the 80s.
By the way, curved traces are still rare in modern EDA software, only rather expensive suites feature it, such as Eagle and Altium.

>>2761606
Just arrived and should fit.
Gonna print a mounting bracket now.

>>2762274
As a kid I found technical drafting by hand on the drafting table fascinating. I even made technical drawings of my own (using normal pen and paper). Unfortunately all of it disappeared, and all is done by CAD programs, which are extremely expensive and only run on Windows. I should try doing some FreeCAD again.

>>2762368
There's KiCAD if you're interested, you can design PCB's and footprints with it. It's FOSS and runs on basically anything, including on Garuda Linux (what I use).

Stopping by with a question.
What is the translucent yellow plastic sheet that can be found on some electronics components? Seems to be intended for keeping solder joints from touching nearby bare metal.

My initial search brings up Kapton Polyimide Film
>There's a variant that's also electrically conductive
Maybe I shouldn't try buying this from generic sites. Are there other names for this stuff? Is this even the right thing?

>>2762368
>FreeCAD
It's 'fine' for small things but an absolute chore once you get into more complicated designs.

>>2762408
>translucent yellow plastic sheet
Kapton tape

>>2762416
Thanks anon.
For some reason, plain non-tape sheets of the stuff are always associated with 3D printing bases.

>>2762405
I meant in general.

>>2762408
>It's 'fine' for small things but an absolute chore once you get into more complicated designs.
It's the only choice. All the professional CAD programs are expensive and locked down with dongles. Nothing a hobbyist could use.

>>2762408
No problem in buying from generic sites.
Kapton is one of those great things that were invented for spaceflight, then seeped through to the consumer market, where it's been a mainstay for decades.

How common are 5-pin micro to micro USB cables? Do they even exist? I bought a random cable and of course it only has 4 wires, no ID wire. I need it for a custom project (not a standard USB connection) where I need all 5 pins. There are plenty of 5 pin receptacles but I didn't realize that 5 wire cables may not be readily available.

>>2762448
>I need it for a custom project
OTG? Do you want to draw current from the host? Put a 100k resistor between sense and ground.

>>2762453
>between sense and ground.
on the host side

>>2762420
No problem, Anon. Kapton is good up to 700F/371C and it's slippery/flat so it's a good use of the material in a 3D printer bed.

>>2762453
No, it is not even for USB. I want to use it as a replacement for a bulky JTAG connector on a J-Link style debugger. I need 5 wires: VCC GND SWIO SWCLK and Vtarget (so that the debugger could sense the target voltage). Worst case, I could add a jumper on the debugger side to feed its own 3.3V as a "target voltage" just so the firmware is happy.
The jlink debugger itself does use OTG but it is a completely different connection to communicate between the PC and the debugger and it is USB-B on the jlink side.

>>2762456
I've never seen a micro USB (USB 2.0) cable with 5 wires, but a type C cable (USB 3.2) has 18 wires.

>>2762458
yeah so I did this. I have some USB-C cables and connectors: they are 16p which is in reality 5 wires. So I can also have reset which is optional but nice to have. I just wanted to add USB-B micro as well "just in case". And I added J6 jumper in case I absolutely have to use a 4-wire cable.

>>2762462
pic for reference

>>2762435
McMaster separates the three variants (insulator, conductive, thermal), but usually Amazon isn't that specific due to the search keyword barf they do. But other sites don't want $18 for one 12x12 inch sheet either. The conductive variant is 3x as costly so probably easy to dodge...

I'll probably just buy a roll of tape

>>2762455
I'm in a bad spot if I hit those sort of temperatures in what I'm using it for. Which is just to cover some bare pcb on a hard disk.

>>2762458
Most micro cables don't, but a fifth wire is actually in the standards document for USB On-The-Go.
It's used as a host/client identifier.
https://www.edn.com/understanding-usb-on-the-go/

>>2762473
Kapton is both an electric and thermal insulator. Chances are those variants are the same or only differ in thickness.

>>2762474
This is tricky, it doesn't actually mention the 5th wire, it is talking about the ID pin being shorted to GND or not. For example the cable I have does have 5 pins but only 4 wires so sounds like the idea is the ID pin is shorted to GND on one side and is floating on the other and that doesn't mean they need to be connected inside the cable.

>>2762480
Things go south, you could try to use the shield braid as a conductor. It's isolated from the other wires but obviously has different impedance.

>>2762482
Hmm I didn't think about that, may be worth trying. This is a weird project already why not make it even weirder. I just realized that I cannot use the ID pin at all since it could be shorted to GND yikes. So adding J6 >>2762462 is a terrible idea as it will result in a short circuit which may potentially destroy the USB port on the PC.

>>2762485
Depending on how weird you want to go, how long it has to be and how much money you're willing to spend, you could go for other types of cables and repurpose them. Most component cables for old consoles, for example, have five data lines (3 video, 2 audio) and shared ground, making six separate conductors.

>>2762489
Yeah I guess the main motivation for this project was to repurpose some cables. I wanted to design a PCB for J-link v9 but there is nothing wrong with my old J-link v8 that used the old style bulky JTAG connector. So I figured I'd add a twist and have some fun. For now I am going to add USB-C as the main connector and an optional micro USB once I figure out what to do about the missing wire.

>>2762448
just use JST connectors like a normal person
or 2x3 IDC if you're a boomer

>>2762456
So low current data line shit?
My dude, any old CAT-5 ethernet cable will more than suffice. Will probably be far more quality than whatever chinkshit USB cables you can buy.

wtf there is a new already? and i was still posting here since it is only page 7. you guys are cruel

>>2762502
I blame tranny zoomers high on their own jenkem.

>>2762496
>>2762501
>IDC
the old one is IDC, too bulky and retro

>CAT-5
bulky and too industrial. it just sounds funny to use CAT-5 for anything other than ethernet.
>JST
those are probably ok but too geeky and arduinish.

USB is super elegant and slick. Plus the cables are really sexy. I have different colors, some bright green ones, some red and one is teal.

>>2762502
If there is they're a retard who didn't post a link to the new thread.

>>2762508
>too bulky and retro
Easy to make your own cables, can be surface-mounted, through-hole mounted, or edge-mounted, the pinouts match common protoboards, they're keyed so you can't put them in the wrong way. 2x3 isn't that bulky.
>it just sounds funny to use CAT-5 for anything other than ethernet
Use CAT-3 instead then.
>too geeky and arduinish
JSTs are industry standard for small keyed connectors. Plus they're 10 times better than using connectors that have protocols already tied to them, you're just asking for confusion. USB-C was a mistake, micro USBs too.

Yeah so some retard made a new thread 3 fucking days ago and never posted it here. Migrate I guess, fucks sake.

>>2761476
>>2761476
>>2761476
THREAD NUVEAU

>>2762549
> better than using connectors that have protocols already tied to them
Overall I agree. I am not 100% serious, I guess I am just bored and want to do something different. But regardless, what is wrong with USB-C or micro-USB?

>>2762553
USB-C has too many different standards tied to it already. It's a mess of PD, QC, Thunderbolt3, Thunderbolt4, USB 3.1, USB 3.2, USB 4, even USB 2 and no data capability at all. And there is no "this port does literally everything" because nobody uses 40Gbps and 200W through the same plug. Now we've got cables that can handle different amounts of power or speeds of data, and have a chip in them that should tell you which one is which. No you can't just read it written on the cable, you need to plug it into a tester or take the manufacturer's word for it on the packaging you threw out months ago. But of course, half the cables you buy say one thing and do another. Some don't have the CC pins properly connected, some have wires too thin to actually handle 3A or 5A or whatever, some are assymetric so it depends on which way you plug them in, it's a nightmare. There's at least two functionally different ways of wiring a USB C male to USB A female adapter. Plus the connector itself has that wafer thin thing in the middle that kids break on their nintendo switch because it's a bad connector for a dock to use. One size fits all bullshit.

Micro USB is just a more finicky, weaker, and harder to solder version of Mini USB. But worse, it's also become the de facto standard for charging things that have a single lithium ion cell in them, even though they have no need for data pins or a low-profile connector. While this is better than generic barrel jacks with unspecified voltage and current limits, the connectors are impossible to waterproof and are frequently the part that breaks first on bluetooth speakers and such. They break often mainly because they're always surface-mounted, but also because they come with 15cm long cables. And those cables are awful, over half the time they have no data pins at all, polluting your cable collection.

If only someone designed a standardised magnetic power connector. A standardised smaller ethernet connector would be nice too.

>>2762568
They had to somehow destroy USB C when they realized this could be the last connector for a while. The other reason for USB C was similar to hdmi’s raison d'être, namely, DRM with active cables containing “protection” devices. As in “protection racket”
We were fine with coax, the bandwidth of that is astounding. People with cablemodems don’t realize everything already comes in through that and they have to buy a new HDMI 3.2 cable every few years to keep up with the Jones's. Again, DRM reasons.

Like the other anon, I suggest plain old telephone cable and RJ11. Some people call this “cat 3” but I don’t think it is. It’s good enough to watch netflix on 1080p whatever it is.
Dirt cheap, reliable, available as salvage with people scrapping land lines for thir starlink facetime.

You can up the voltage to, say, 48 V to effectively get more power through it and buck-convert it down to whatever you need. 48 V is a telecom standard and most equipment on telephone lines is protected.

>>2762568
They had to somehow destroy USB C when they realized this could be the last connector for a while. The other reason for USB C was similar to hdmi’s raison d'être, namely, DRM with active cables containing “protection” devices. As in “protection racket”
We were fine with coax, the bandwidth of that is astounding. People with cablemodems don’t realize everything already comes in through that and they have to buy a new HDMI 3.2 cable every few years to keep up with the Jones's. Again, DRM reasons.

Like the other anon, I suggest plain old telephone cable and RJ11. Some people call this “cat 3” but I don’t think it is. It’s good enough to watch netflix on 1080p whatever it is.
Dirt cheap, reliable, available as salvage with people scrapping land lines for thir starlink facetime.

You can up the voltage to, say, 48 V to effectively get more power through it and buck-convert it down to whatever you need. 48 V is a telecom standard and most equipment on telephone lines is protected.

>>2762568
Most of these advanced protocols are just tunneled. The only thing that needs to be done is switching between the protocols. And teaching the user that he can't have two high bandwidth protocols at the same time. (Like USB-3.x-connected displays that only do USB 2 on the builtin USB hub.)

>>2762619
>DRM
DRM doesn't need active cables. New HDMI cables are only needed to accommodate higher bandwidths. You can do DRM at any bandwidth.
>48 V
I know nothing about electronics (hi /ohm/) but apparently that's not ideal either Personally I'd just have said keep the power line always 5V, discard the CC lines, and add a separate 48V line needed only by high power shit like USB-connected vacuum cleaners and power banks. But could that work bi-bidirectionally? (With USB-C, the PD protocol actually negotiates the role a device takes. Power banks can switch roles dynamically.)

>>2762619
the jew fears the self-crimped cable

>>2762568
>>2762619
Imagine a perfect world with magnetic connectors. Waterproof, dustproof, easy to clean and it takes active effort to break them.

>>2762858
They better not be as shitty as apple's magwhatevers.

>>2762858
My zojiroushi electric kettle has magsafe connectors. Been working fine fine for 20 years, used all day, every day.

The reason is safety, if you put you kettle (or laptop) on a table and somebody walks through and trips on the cord, it won’t go flying.

And no, apple didn’t invent it as far as I’m aware.

>>2762858
That connector is hilarious.
> universal SERIAL bus
At most, this needs only 4 wires, but you can certainly get by with only two and put a bias voltage on the data lines like satellite antennas.
> b-but my new phone needs more power
No it doesn’t. Battery technology peaked years ago and most phones have Qi charging like my toothbrush which is actually waterproof, dust proof, etc.

>>2762935
>this needs only 4 wires
>high-speed data, network, audio and video via a single twisted pair
Read this before spouting nonsense:
https://www.usb.org/sites/default/files/D1T1-2%20-%20USB%20Type-C%20System%20Overview.pdf

>>2762959
>high-speed data, network, audio and video via a single twisted pair
That is possible.

>>2762959
I mean serial in general, not the frankenstandard du jour.
The retards at USB central didn’t invent serial communication.

>>2762858
The only one I know about is apfail'/s and it's terribad. I'm really glad that work provided me with a mac laptop so I could see how mind blowingly amateurish the OS is and how cheap the hardware is despite the insane pricetag.