[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/diy/ - Do It Yourself


View post   

File: 776 KB, 751x942, 1532228105103.png [View same] [iqdb] [saucenao] [google]
1606134 No.1606134 [Reply] [Original]

generally done thread: >>1600531

>RULES
0. Electrics ≠ electronics. Home appliances/sparkies to /qtddtot/ or /sqt/. PC assembly to >>>/g/.
1. Do your own homework. Search web first. Re-read all documentation/datasheets related to your components/circuits. THEN ask.
2. Pics > 1000 words. Post relevant schematic/picture/sketch/9001.5 hours in MS Paint with all part numbers/values/etc. when asking for help. Focus/lighting counts.
3. Read posts fully. Solve more problems than you create.
4. /ohm/ is an anonymous, non-smoking general.

>I'm new to electronics, where to get started?
It is an art/science of applying principles to requirements. Find problem, learn principles, design and verify solution, build, test, post results, repeat

>Project ideas:
http://adafruit.com
http://instructables.com/tag/type-id/category-technology/
http://makezine.com/category/electronics/

>Principles (by increasing skill level):
Mims III, Getting Started in Electronics
Platt, Make: Electronics
Geier, How to Diagnose & Fix Everything Electronic
Kybett & Boysen, All New Electronics Self-Teaching Guide
Scherz & Monk, Practical Electronics for Inventors
Horowitz and Hill, The Art of Electronics

>Design/verification tools:
LTSpice
falstad.com/circuit/circuitjs.html
NI Multisim
CircuitLab
iCircuit for Macs
KiCAD (pcb layout software, v5+ recommended)

>Components/equipment:
Mouser, Digi-Key, Arrow, Newark, LCSC (global)
RS Components (Europe)
eBay/AliExpress sellers, especially good for component assortments/sample kits (caveat emptor)
Local independent electronics distributors
ladyada.net/library/procure/hobbyist.html

>Related YouTube channels:
mjlorton
jkgamm041
eevblog
EcProjects
greatscottlab
Photonvids
sdgelectronics
BigClive

>Li+/LiPo batteries
Read this exemplary resource first: https://www.robotshop.com/media/files/pdf/hyperion-g5-50c-3s-1100mah-lipo-battery-User-Guide.pdf
>I have junk, what do?
Take it to the recycler.

>> No.1606143

>>1606134
>pic related
cmon, don't leave us hanging, what was the cause?

>> No.1606147

>>1606143
>By Saturday afternoon, City Councilman Chris Glassburn announced that the mystery had been solved:
>The source of the problem was a homemade battery-operated device designed by a local resident to alert him if someone was upstairs when he was working in his basement.
>It did so by turning off a light.
uuh
>“He has a fascination with electronics,”
>Mr. Glassburn said, adding that the resident has special needs and would not be identified to protect his privacy.
oh no

>> No.1606160

>>1606147
It was persistently putting out a 315 MHz signal. He was lucky a local ham investigated and not the FCC.

>> No.1606174

Hello my dudes, I'm about to embark on a 25000 kilometre motorbike trip and I am looking for a tiny multimeter to fit in my tool roll. All it need to be able to do is vdc, resistance and continuity. Is there such a thing? I found one that looks ok but it connects to your smartphone and that's just one more thing to go wrong.

>> No.1606176

>>1606160
>continuously puts out a signal
yeah, that's not how you're supposed to Part 15

>>1606174
the $2-if-not-free pocket meters from Hazard Fraught should be enough

>> No.1606182

>>1606176
That's still too big senpai, will fuck up my tool roll size and won't be able to fit it in my nigger rigged PVC pipe system. I'm talking like, iPod nano size.

>> No.1606184
File: 148 KB, 700x591, FS1000.jpg [View same] [iqdb] [saucenao] [google]
1606184

>>1606160
Now we know how to wreak havoc on the neighborhood and that an effective quarter wave antenna has a length of 24cm. Shop with confidence.

>> No.1606222
File: 77 KB, 800x800, 1555751238150.jpg [View same] [iqdb] [saucenao] [google]
1606222

>>1606182
so like this?

>> No.1606226
File: 122 KB, 530x530, 1529820837194.jpg [View same] [iqdb] [saucenao] [google]
1606226

>>1606182
or?

>> No.1606229

>>1606182
How about: three wires, a 9V battery, and your tongue?

>> No.1606231

>>1606226
That's the ticket, thanks senpai.

>>1606229
Kek

>> No.1606240
File: 112 KB, 1211x649, 1552898745951.jpg [View same] [iqdb] [saucenao] [google]
1606240

>>1606134
this thread's digits are brought to you in part by the Stancor P-6134 power transformer

>> No.1606256

>>1606240
>a transformer
I don't need to nitpick, but isn't there something a little more "active" you could have picked? It just doesn't feel terribly exotic. Especially since there are plenty of TI products in the 6XXX range. Namely, a quad rail-to-rail amplifier.
Or were you thinking we were getting kind of sick of one op-amp or switching converter controller after another? In which case, I wonder if there are any vacuum tubes that we might be able to use for the next thread? Or other more retro parts, like paper-oil capacitors or trimmer capacitors.

>> No.1606271

>>1606222
cute, wonder if it's as decent as the 8008

>> No.1606284
File: 16 KB, 474x474, 1538532682907.jpg [View same] [iqdb] [saucenao] [google]
1606284

>>1606256
what's that? you think you're too GOOD for power transformers, boi?
just you wait until we get quad zeroes

>> No.1606288

>>1606284
Well, to be fair I own both.

>> No.1606310

>>1606288
ah, well that's fair enough
>terribly exotic
getting there tho, a 6.3Vct trafo just doesn't see much use outside of a coffee maker or tube filaments
>we were getting kind of sick
mostly I was just in the mood for a little change. kinda like a time long ago when I chose a boring old hex spacer because reasons such as laziness and being in a hurry

>> No.1606345

>>1606310
Yes but is that 6.3V RMS or peak? If it's RMS then it has a peak of about 9V, which is slightly less exotic.

>> No.1606356

Anyone have any good beginning books on circuit design? Does this section have a wiki I could check out? I'm familiar with basic signal processing concepts, but I'm only familiar with the digital side of things and have no idea how analog works

>> No.1606358

>>1606356
Google 'agarwal Foundations of Analog and Digital Electronic Circuits pdf'
It's the book that goes with the MIT 6002x online course, well worth taking as an introduction to electronics.

>> No.1606369
File: 10 KB, 400x400, bike lock.png [View same] [iqdb] [saucenao] [google]
1606369

>quiescent current: 0.1µA
>operating current 20mA
Boy is that a big difference. The q current is so low that I could run this thing for a year on a 1F supercapacitor, but only a few minutes when woken. Which would be fine if I wasn't going to wake it for a while, so I guess I'll be going for a couple NiMH cells.

Bike lock designing anon here, after looking into how to set the MCU to sleep to preserve battery, I found that you basically have to do it with an external interrupt. So without an RTC or something similar burning my power, it would be impossible to check continuity every second or so. Hence instead I've decided to go for my FET-buffered continuity sensor, pic related.
R0 is the only resistor that current is flowing through in standby, so I want to minimise this, for which using one of my 10MΩ resistors should work fine, provided the MOSFET's gate plays nicely in that instance. I could possibly use a JFET if that would make things easier. I've tested this part of the circuit with a 1MΩ and it works fine.
The switch on the right is a key switch that goes in the end of the handlebar that disables the entire system, while the two wire loops come out of the other end. The batteries will be in there too. Not pictured is the alarm, which will need to be somewhere protected but also audible, and the charging system, which will basically be an inductor, rectifier, filter cap, and 5V regulator, with a magnet tied to a spoke to pass near the inductor on the fork.
It's pretty hard to test whether my off-the-shelf inductors and magnets will pack enough punch to get a decent voltage off a magnet, so that will have to wait until I have a bike to test it with.

>> No.1606370

>>1606134
Soooo, you saying someone could go on ali, order a bunch of cheapo antennas, make battery powered transmitters with his TS100 and hide them around the neighborhood for shits and giggles?

>> No.1606374

>>1606370
Easily. But you could be more disruptive than that. Normal hobby transmitters use the MHz ham radio bands, which are mostly used by non-critical things. So by changing your transmitting frequency to one that wifi, or cellular networks, or GPS uses, you can likely cause far more havoc.

>> No.1606376

>>1606374
You will need a much more expensive setup with a big battery to jam wifi or cellphones since unlike keyfobs they have big ass antennas

>> No.1606377

>>1606376
You'd just need to send out extra erroneous blips and bleeps, not necessarily stronger signals.

>> No.1606380

So I heard about the Basic Engine ( https://basicengine.org/ ) and thought i'd be a fun, cheap project to learn with (not shilling). They advertise that it can be build for around $10, but none of the parts I can find are anywhere near that cheap on aliexpress and all the other sources.

Is this just bullshit?

>> No.1606381
File: 36 KB, 417x403, vHg6E.png [View same] [iqdb] [saucenao] [google]
1606381

>>1606369
Ok, anyone have experience with NiMH charging setups? A trickle-charge circuit probably won't work since I'll be using the bike for anywhere from 10 to 40 minutes a time, so I might be looking at negative voltage gradient sensing if my current level is high enough. If the current is low enough (or I want to sacrifice some charge for peace of mind), I may be able to approximate the voltage I'm aiming for and do a hard-cutoff with a schmitt-trigger comparator. Anything more complicated than that will lead to me using the MCU for that also, which isn't necessarily bad, but I'll be using an ATTiny85, and the less program space I have to waste the better. A lookup table of a bunch of current-voltage values might be out of the question, but if not an approximate mathematical relationship would do the trick. I might be doing a fair bit of battery testing in that case.

>> No.1606389

>>1606381
Actually I can probably set the end voltage to 1.35V or so, then set an LM317 to that (actually to 2.7 since there will be two cells in series) and limit the current with a resistor, and end up with a charge that starts off quick, and as each cell approaches this 1.35V, at high currents, the current will automatically drop off until it trickles to a halt at the desired voltage. It's not efficient since I'd be wasting probably a majority of my power in the resistor and LM317, but it would give me the freedom to not bother with any fancy charge circuitry. I could also go for a single lithium ion cell with similar circuitry if that would be better.

>> No.1606406
File: 1.25 MB, 800x600, file.png [View same] [iqdb] [saucenao] [google]
1606406

the power switch for my hammond maverick fucked off the other day. the red plastic flip thing just crumbled beneath my manly finger and exposed the inner workings of the switch: just two metal plates where if they connect, the organ gets power.

i wanna change the switch out so i bought one rated 6A/125VAC or 3A/250VAC. i'm in europe so mains is 230 i believe. so, do yall think that this damn thing draws less than 3A?

i've googled but very little is written about this particular model because its a piece of shit. i did however find out that they have 200 V capacitors in here that supposedly take a day to discharge after unplugging it. what the fuck

>> No.1606407

>>1606406
Get a kill-a-watt and see. Nothing wrong with shorting the metal contacts together or just replacing the broken switch with a thick enough wire for testing. I wager it pulls significantly less than 3A because that's the better part of 1kW, and that's a lot of heat to burn.

>> No.1606410

How do i use that pink soldering liquid?
I wanted to solder a wire to an aluminium tape so i dripped the liquid on it, put the wire there and started melting solder on it with my iron
the liquid boild and vaporized leaving brown burn marks behind but the wire was not soldred
also the vapors made my head hurt unlike just regular solder fumes, weird

>> No.1606412

>>1606410
Can't solder to aluminium. Also that pink liquid sounds like something for pipe soldering, in which case you'd be better off asking in /sqt/ or some plumbing thread. Hey wouldn't it be funny if we had a "national pipe thread"?

>> No.1606416
File: 120 KB, 1238x682, digispark.png [View same] [iqdb] [saucenao] [google]
1606416

I'm taking a look at the digispark board here, because their feeding USB directly into two pins on the mcu is interesting. There's nothing about USB compat on the datasheet, and they're using some interesting circuitry, namely those zener diodes on the data pins, one of which also has a pullup.

I was thinking about making a dongle like the normal digispark, but without the massive 5v regulator, and with the dip8 socket like other digispark style boards. I'd make it the same width as the usb plug all the way along since my digispark has too short of a plug to work in my laptop's recessed usb sockets, and I might even ditch the pin headers and just use it for programming dip mcus. I'd call it the digidong. Thoughts?

>> No.1606418

>>1606147
>>1606160
I suspect they'd have taken the same tack as Mr Glassburn.

>> No.1606422

Whats the safest way to use 400V (low load) on a raspberry pi hat board?

>> No.1606424

How much force does it take to mill copper away from pcb?
Will i fuck up my 3d printer if i mount a small motor with a mill bit on it?

>> No.1606425

>>1606406
Does it have a fuse?
What value (amps) is the fuse?
The switch should be of equal or greater capacity.

>> No.1606433
File: 62 KB, 709x472, Olive-oil-flux.jpg [View same] [iqdb] [saucenao] [google]
1606433

>>1606412
>Can't solder to aluminium.
It's not the alu, it's the oxide.

>> No.1606435

>>1606433
Well I doubt it would wet too well to the metal itself if you were wanting a structural join, but fine for electrical work I guess. Still, he said "aluminium tape" so I doubt it would be too easy to just sand away the oxide without fucking up the tape.

>> No.1606444

I have a battery powered device.
As the battery discharges the voltage drops.
The voltage is between always 5V and 3V. I can use LDO between 5V and 3.5V but once the battery voltage hits 3.5V i have to bypass the LDO and feed the battery power directly into my device.
What would be the easiest way to do this?

>use better ldo
not an option

To put it more simply, if the voltage is >=3.5V send power into ldo (which then sends it into the device), if <3.5V send power directly into device.
So far my solution is to use attiny85 to measure the battery voltage and depending on the voltage it either turns on mosfet A or mosfet B to route power into the LDO or directly into device, but that seems overly complicated to me

>> No.1606446

>>1606407
sounds a bit like danger zone, and i've got no mojo for this shit. it'll probably come to it anyway. thanks for the measuring tip

>>1606425
ah right, thanks! seven fuses. i remember changing two of them once. previous owner/manufacturer had stapled an envelope with the correct values to the inside of the thing.

>> No.1606447
File: 281 KB, 489x321, cu-al.png [View same] [iqdb] [saucenao] [google]
1606447

>>1606435
No sanding, use tip in oil.
https://www.youtube.com/watch?v=_mYkM9lHMho

>> No.1606456

I accidently reversed polarity of lithium battery charger (ebike). Is it fucked?

>> No.1606457

>>1606456
I haven't disassembled it yet. What should I expect? Blown diode with fuse? Or something worse?

>> No.1606461
File: 30 KB, 1533x1234, q8i5r3.gif [View same] [iqdb] [saucenao] [google]
1606461

>>1606457
>>1606456
I think I found schematic. Only rectifying diode and 0.1 ohm resistor would die, right?

>> No.1606470

Got a bunch of 1n34a germanium diodes from aliexpress. They look real enough. Glass DO-7 package and all.
Like 2 or 3 of them show the expected Vf of around 0.35V but the rest is all over the place, ranging from 0.5V to 1V. A couple even show a reverse bias voltage of around 2.7V on my DMM (although come to think of it, that does make sense given their high leakage current, which might make the DMM think it's just measuring forward bias at whatever current it uses to measure Vf at, and not an indication of a defective diode).
Is this kind of variation in Vf normal in Ge diodes or are these just particularly shit?

Guess I'll never learn my lesson buying chinesium crap.

>> No.1606503

>>1606461
Okay, I broke diode bridge and transformer while I was disassembling shit. Fun...

>> No.1606556

>>1606503
Nah, fuck this piece of shit. I'm getting new one.

>> No.1606569
File: 299 KB, 790x870, Ge-Diodes.png [View same] [iqdb] [saucenao] [google]
1606569

>>1606470
>variation in Vf normal in Ge diodes..?
Vf is often poorly specified (like < 1V @ 5mA), but may not vary that much within the same lot. Are your diodes marked or just glass? I have no 1N34 but some other AA.. types. The six AA144 from the same tape I checked had 255±3 mV at about 1mA. For comparison: BAT43 250 mV, 1N5819 200 mV, 1N5817 160 mV, AA135 190 mV, AC152 B=C ≈100mV, all at about 1mA. Pic: top high R bottom low R.

>> No.1606573

>>1606569
Just glass, no markings. They look almost the same as some legit ones I have, except the glass is crooked as all hell, lmao. Think they might just be shit. Like most anything the chinks sell. Still, at €0.13 a piece I kinda half expected to get the real deal. Have some more coming in the mail from ebay, hope those are better. Otherwise it's back to buying them from Tayda for €0.28 each.

>> No.1606574
File: 520 KB, 1574x1500, ninebot.jpg [View same] [iqdb] [saucenao] [google]
1606574

>>1606556
This one? Not cheap.

>> No.1606583

>>1606574
Ninebot one was pretty good in terms of schematics safety and such. I don't think I would get this from cheap chink, but wherever.
I really need to learn how to disassemble those sealed cases.

Next time I will add fuse and diode, since my connector is naughty and sometimes I can insert it wrong easily.

>> No.1606586
File: 1.29 MB, 3264x2448, not_mine.jpg [View same] [iqdb] [saucenao] [google]
1606586

So I did a MIDI controller prototype with some buttons and potentiometers using a UNO and HIDUINO.

You basically reprogram the firmware on the USB IC on the UNO with HIDUINO and it will translate MIDI to USB signals without needing translating software on the PC.

Now I want to do my own PCB for a better MIDI controller with more buttons and potentiometers so I thought of using a mega since it has more analog inputs but it's too hueg for my case. I saw this Mega pro mini which is much smaller, problem is it it has no USB IC, which I need for the HIDUINO.

So I thought of adding a pin header for the mega pro mini on my pcb and wire together the serial communication to another atmega (like the one for leonardo for example) and use that one as USB IC. Theoretically it should work, but has anyone tried something similar? Officially HIDUINO supports only UNO and MEGA because of this USB-IC thingy.

I've seen other MIDI libraries out there for atmel chips with integrated USB communication but I really liked this one.

>> No.1606599

>>1606583
>need to learn how to disassemble those sealed cases
https://www.youtube.com/watch?v=a00esZ3B-aQ

>> No.1606602

>>1606599
>applying gasoline to electronic components
>that tend to get hot
>having a gasoline soaked cotton ball
Are you sure there isn't a better way to do this?

>> No.1606608

>>1606416
anyone?

>> No.1606609

>>1606602
Other solvents may work too, like acetone or ethanol.
Method not suitable for over-protected children who are rumoured to cause four out of three housefires because they have been denied the ability to acquire common sense.

>> No.1606622

>>1606599
I tried acetone, it usually opens that shit, but Ninebot charger was stubborn. I guess gasoline has some sort of acetone in it, and unlike pure acetone it doesn't evaporate that quickly.

>>1606602
Gasoline is just as flammable as other solvents, so yeah. Also, it evaporates quickly.
>>1606609
Alcohol (IPA or tasty) doesn't work.
I will try next time mixing acetone and white spirit or mineral thinner...

>> No.1606690

>>1606422
use a relay

>> No.1606706

>>1606380
~$10 burgers excluding shipping for just components seems about right. Most expensive bit is the video chip, https://www.tme.eu/en/details/vs23s010d-l/ $5.1 + shipping, ESP-12E ~$1.6 from china https://www.aliexpress.com/item/the-s-stands-for-security/32983867729.html , get all the passives from LCSC, for pcb manufacture check pcbshopper.

>> No.1606708

>>1606389
For easy constant voltage + either constant resistance or constant current Li-ion is the way to go.

>> No.1606710

>>1606422
Relay.

Okay, how do I modify e-bike controller to engine brake? Not regen, but braking, heating resistor or something.

>> No.1606729

>>1606710
Brushless motor? Can you change/replace the control software? If so, best way is to regen then use constant voltage load circuit to dump the excess.

Brushless and don't want to change the controller? 3 phase rectifier made from 6 diodes, then dump the desired amount of power on the DC side, can do resistor + any switching device. Can do linear transistor/MOSFET or PWM it to adjust the amount of braking.

>> No.1606739
File: 78 KB, 746x412, www.png [View same] [iqdb] [saucenao] [google]
1606739

>>1606710
Why pic related will not work? Or will it?

>> No.1606741

>>1606729
>Can you change/replace the control software?
No, unfortunately, and I'm too retarded to make my own. Driving brushed DC motor was tricky... I burnt 15 MOSFETs for sure. BLDC is fucking magic for me.
> If so, best way is to regen then use constant voltage load circuit to dump the excess.
I can use regen, but it kills chinkshit batteries pretty fast, and it provides no benefit (we are speaking about 50-300 meters of saved charge, that is nothing)
>Brushless and don't want to change the controller? 3 phase rectifier made from 6 diodes,
That is like 5-15A of current, if not more. That is 2.5-7.5W. Diodes with such dissipation are expensive. So, this is not an option.
Infinitive adjustment is not needed, so I think I will be fine with two levels of braking, full retard, and retard. Thus, I can use relay for this. >>1606739.
Relays are cheap, resisors are cheap too, steal some wire, cast it in gypsy, and here is your 900W resistor.

>> No.1606743

>>1606739
Oh, +5 is on top of hall sensors, and ground is under.

>> No.1606747

>>1606708
Right, thanks. If I can use a single LM317 to do both the CC and CV I'll do that, else I'll just go with the resistor. Actually I have some JFETs that I could use to make a constant-current "diode", but I bought them for RF purposes for like 80c each so it might be a waste.

>>1606710
Usually you'd just short the motor windings and have the heat build up in the motor itself. Any motor is likely going to have a higher dissipation rate than all but the beefiest of power resistors.

Anyhow my chinky 8 channel USB logic analyser arrived, anyone else have one?

>> No.1606748

>>1606747
>Usually you'd just short the motor windings and have the heat build up in the motor itself.
Hm... It is a big chunk of metal, so yeah. But I think in this case relay will become a resistor, especially chin-chong ones.

>> No.1606750

>>1606748
Oh yeah that's true. The moment you start doing impedance matching calculations with a relay you need to back away and re-think.

>> No.1606758

>>1606750
Anyway, I will try relay madness with steel (or nichrome) resistor >>1606739 I designed. Maybe it will work, who knows.

>> No.1606759

>>1606747
Dead short across ebike motor is gonna be too harsh.

>>1606741
Suitable diodes aren't that expensive, keep in mind they're rated by mean current and you're not going to be running 100% duty cycle, something like this would be more than enough https://www.digikey.com/product-detail/en/smc-diode-solutions/MBR1645/1655-1033-ND/6022123 Do kinda want schottkys so that you're not burning much power in reverse recovery when the controller is doing PWM forwards.

But yeah, 2 level pure resistive with relay control should work okay, if you don't mind some vibration could do the arrangement there with only one relay clicking on for effectively single phase low rate then second relay for full wye braking.

>> No.1606763

>>1606759
>if you don't mind some vibration could do the arrangement there with only one relay clicking on for effectively single phase low rate then second relay for full wye braking.
Well, I will not notice those vibrations, due to kick scooter (which I called e-bike for some reason, lol) is vibrating like like a dildo by itself. Sound will double as ring.
>Do kinda want schottkys so that you're not burning much power in reverse recovery when the controller is doing PWM forwards.
They use black magic, they turn on fets in appropriate moments, so resistance is low. In factory-made controllers.
Simple diode will work (after all I dissipate heat anyway), but I have idea how to cool it.

>> No.1606793

>>1606586
there is an arduino general that deals with that platform and may have more pertinent advice for you. personally, on a lazy day, I'd just treat the duino board as a reference/example circuit and incorporate as much of it as needed onto my board

>>1606608
complete project and come back with pics. we love it when people do that

>> No.1606800

>>1606706
are the passives cheaper from LCSC than aliexpress? I can find all the components on Ali except the video chip, but I have to buy 10 each and that kinda drives the price up to around $15 per unit (assuming I built 10 of them)

>> No.1606802

>>1606800
LCSC is nicer than buying from separate sellers on Aliexpress since you'll get one package with decent labelling and cheaper shipping

>> No.1606825

There is a $3 handling fee since the order is below 15.. Anything that's useful from LCSC I should add for possible future projects for learning? Or mayble console modding, fixing vintage PCs, etc...

>>1606802
you're right, they're actually cheaper than Ali after I checked (except for some components they don't carry). Thank you anons

>> No.1606841

>>1606793
I was more asking what people thought of the zener diodes, but thanks for the encouragement. It shouldn't be terribly hard to design either, but I've no clue how much on-chip memory needs to be taken up by the "USB programmable" code, which would determine how worth making the product would be.

>> No.1606842

>>1606825
CH551 one is neat, it's a 20 cent MCU that can do USB.
And the CH554 is a 70 cent MCU that can do USB Host.

>> No.1606844

>>1606842
Oh and they have cheap breadboards

>> No.1606859

I want to to make my own pcbs, but the toner transfer method is gay as fuck and imperfect as hell.
So i am going to use a 3d printer insted. I can either glue a sharpie to it and draw the schema on the copper, but then i have to fecl3 it and fecl3 is gross and hard to get rid off.
Or i can glue a small motor to the printer and mill the copper off instead. This completely avoids the acid and leaves perfect 100% pristine undamaged copper traces, but it makes a lot of mess and copper and shit fly everywhere, plus you have to anchor the pcb much better for milling than for sharpying

so i'm not sure which will be better

>> No.1606861

>>1606859
>but then i have to fecl3 it and fecl3 is gross and hard to get rid off.
You can use etch system from Seno, it keeps the acid chamber always closed. It's like 10$ or you can even make it yourself for less.

>> No.1606866

>>1606859
mill can also make the holes for the tht components so that is a huge advantage. having to drill 100 holes by hand will make you want to kill yourself

>> No.1606869

>>1606739
What relay are you using?
Relay are not attend to control motor, correct me if I am wrong but it has a certain lifecycle and controlling a motor with 2 open close by rotation.... if you have a 1000 rot/min, it won t last for days.
You can look at the datasheet of the relay.

I would rather use mosfet with flybakc diode instead of relay. Or a simple ln293. But what kind of motor you want to control?
It seems like a brushless but maybe I am wrong. If it is forget it and take a continuous current unless you know what to do.

>> No.1606871

I'm new to picking parts. I need this one:
https://www.digikey.com/products/en?keywords=91A1A-B24-D15L-ND

But it's (relatively) expensive as fuck. Is there a functionally identical one with the same specifications?

>> No.1606874

>>1606866
>tht
>current year

>> No.1606875

>>1606869
The relay isn't going to be switching with motor rotation, just on/off when braking is desired. They're not motor driving, just brainless braking.

>> No.1606876 [DELETED] 

What's it called when a capacitor is shaped like this? With leads coming from the top and bottom like this. Can only find them with both leads coming out of the bottom

>> No.1606877

>>1606874
as opposed to what? the smd fad? no thanks

>> No.1606880
File: 19 KB, 640x640, cap.jpg [View same] [iqdb] [saucenao] [google]
1606880

What's it called when a capacitor is shaped like this? With leads coming from the top and bottom like this. Can only find them with both leads coming out of the bottom

>> No.1606882

>>1606880
axial

>> No.1606883

>>1606880
Axial. Two leads the same side (normal) is radial.

>> No.1606887 [DELETED] 

>>1606880
axial, not radial

>> No.1606888

>>1606859
Why not use transparencies and photoresist? Bonus points for making a lightbox, which is a neat little project. Pretty sure those 405nm UV LEDs work for that purpose.

>> No.1606889

>>1606871
>plastic log
did you ask in /b/
seriously though, you need a 10kohm log taper PC mount pot? some specs are important and some aren't, you have to compare on the points important to you. e.g. dimensions are important if on a crowded board or panel, power handling might be important, lifetime might be important to you, element type is sometimes important but I couldn't tell you the difference between them, etc. maybe the 45 cent Alpha is good enough for the panel this is going into, maybe it isn't?

>>1606877
>he thinks it's a fad
smells like Geritol and Monster Zero in here

>> No.1606892

>>1606880
To add to the other responses, it's fairly common for crossover capacitors to be axial, so if you're buying a replacement part for a speaker circuit, you should search for a crossover capacitor specifically. If not, I'd still check whether the part you want needs to be non-polar, low ESR, etc.

>> No.1606893
File: 35 KB, 640x640, AR215F334K4R.jpg [View same] [iqdb] [saucenao] [google]
1606893

https://lcsc.com/product-detail/Multilayer-Ceramic-Capacitors-MLCC-Leaded_47nF-473-10-50V_C94710.html
and
https://www.digikey.com/products/en?keywords=478-5118-ND

Are these both basically the same part?
Capacitance on one is listed as 0.047µF and on the other it's 47nF, and I can't find a 0.047µF one anywhere on LCSC or Aliexpress. Is "47nF" just the way they write it or is it just an obscure part?

>> No.1606894

>>1606889
it's fad for the dyiers

or tell me, how exactly do you plan on soldering that .5mm long resistor onto that board with your shaking hands and no $1000 microscope?

>> No.1606895

>>1606893
1000 nanos = 1 micro
they are the same

>> No.1606896

>>1606893
learn your SI prefixes bro
like the difference between m and M

>> No.1606899

is Tolerance inportant in selecting a ceramic capacitor?
For example, can I use a ±5% part in place of a ±10% one if all other specs (V, capacitance, etc) are the same?

Thinking of using an MMER102J100V inplace of a AR211C102K4R

>> No.1606901

>>1606899
nah you always round down/up to a multiple of 10 or 47

>> No.1606903

>>1606899
Yes it's fine to use smaller tolerance caps in place of larger tolerance ones. But ESR and ESL are stats that may be important, depending on what your cap is being used for.

>> No.1606907

>simple arduino blink program takes up 700B program memory
>ATtiny13 has 1kB program memory
That's a bit close for liking, how low can this get with a professional development environment?

>> No.1606912
File: 59 KB, 379x672, Clipboard01.jpg [View same] [iqdb] [saucenao] [google]
1606912

what the fuck?
why do they all have different voltages?

>> No.1606913

>>1606907
It's the libraries that take space.
When i upload my code that has 2 lines to esp it takes 53% of memory, if i upload my code that has 200 lines of code, it takes 54% of memory

>> No.1606915

>>1606894
No DIY electronic kit is complete without one of these babies
https://www.youtube.com/watch?v=tn0EKtLOVx4

>> No.1606916

>>1606912
Because they're using different materials with different properties to emit different light.

>> No.1606917

>>1606916
I don't see colors, only photons.
One beam, united.

>> No.1606919

>>1606917
Sorry but you need to be at least 625nm tall to make a girl red, wavelet.

>> No.1606950

>>1606912
Photon energy is proportional to frequency. The energy of the photons is equal to the bandgap voltage times the charge of an electron. The energy varies from 1.65 eV for red light to 3.1 eV for violet.

>> No.1606953

How do you filter a noisy 5-30uV signal? I dont want the distorsion to be amplified but I guess I have to accept that its too weak?

>> No.1606954

>>1606953
For a radio 5µV is a lot.

>> No.1606961
File: 3 KB, 160x128, 1480723571992.jpg [View same] [iqdb] [saucenao] [google]
1606961

>bought €100 worth of stuff from Tayda last month
>they post another 15% discount coupon
>urge to buy more intensifies
Might as well be buying crack

>> No.1606962

>>1606961
but..but the Shopping Experience!

>> No.1606972

>>1606869
Relays will be used only for braking.
Acceleration is handled by controller.
>But what kind of motor you want to control?
BLDC, 250W motor. I push 15-20A in it (I don't have a multimeter, that can measure such currents)

>> No.1606996

Pulled a massive transformer from an old audio amp. Still works and I want to use it in a project I have, but the output is about 10Vac higher than I want.

Can I just unwind some of the secondary coils until I get the voltage down? I've heard of doing this before, but never attempted it.

>> No.1607003

>>1606996
>some of the secondary coils
That can be done. How many coils does it have? If one happens to have 10V you could subtract its voltage from the one you want to use.

>> No.1607026

>>1606996
Yes.

>> No.1607045
File: 1.56 MB, 2048x1675, istockphoto-1131535555-2048x2048.jpg [View same] [iqdb] [saucenao] [google]
1607045

Fucking max ratings, how do it work???
I'm always confused on how absolute max ratings like power or current are determined. For example:
DO-41 (glass) zeners are all rated for 1W, can you really run them >150 C for any appreciable amount of time?

Furthermore, I'll notice often on opamps, they list say +-22V, with indefinite short current of 25mA, but they almost always say "don't do this -> damage to device". Why would they have chosen that limit value then, vs say 10mA (eg. Pd < 500mW worst case). Not to mention, these specs seemingly stay unchanged if they port to a dual/quad package. idk its annoying

>>1606961
other than some sketchy hardware that I cannot recommend(really ass dip/sip sockets, tiny ass headers w/ 50% chance of no connect, etc) they are a godsend for anything discrete or old schoolish IC (cheep 2). I was able to custom pick maybe 5000 1% resistors while still cheaper than chink-bay kits.

>> No.1607075

>>1607045
>I'm always confused on how absolute max ratings like power or current are determined.

Any rating that ultimately has to do with heat is basically just the manufacturer's best guess. Or basically just outright bullshitting in some cases, like rating FETs by the capacity of the die itself (which, for higher-power/low RDS-on models, tends to end up in ratings far beyond what even the package leads can handle without melting).

Simply, a maximum die temperature is established, thermal resistance between the die and package/heatsink is determined, and then a maximum current is derived based on that maximum allowable temperature and thermal resistance. Obviously, this means there are a lot of assumptions made based on how the part is actually installed. Usually favorable ones, as that makes the part look better. Realistically, the one doing the design for the final assembly is the one who has to determine what kind of power they can expect to safely get out of a part, based on given specs, airflow, duty cycle, etc.

The listed power/current rating is just a guideline, really.

Voltage is another matter, as that's usually determined by the breakdown voltage of a PN junction, insulation layer, air gap, or something else that can be known fairly accurately and can't easily be exaggerated without the manufacturer getting into trouble for almost-but-not-quite lying about the capability of their parts. Unlike slow degradation or delayed failure due to thermal stress, over-voltage failures almost always cause immediate and catastrophic destruction of a semiconductor. Passive components tend to fare better, but tend to have other issues if pushed too hard on the voltage front.

In any event, just remember that the datasheet is just that: A sheet of data. Not law.

>> No.1607090

Can I connect couple laptop power supplies to get higher voltage?
>inb4: disconnect ground

>> No.1607105

Where the FUCK can I get T18 tips in the UK
Chinkland doesn't even sell any except for like one listing with no reviews, and I found one distributor which asks you to write them an email with your order and every tip costs £6-7. Fuck this

>> No.1607108

Anyone know max current draw from a typical microphone jack? An electret uses ~.5mA, but some info puts it as high as 1.5mA. If that's the case, I could probably draw 1mA for an amplifier, and get +20-30dB without a separate power source. I already have a design simulated for a class A that'd work, but I'm curious on previous experiments with this.

>> No.1607110

>>1607045
I tied the outputs of an lm358 together assuming they were matched enough for it to not matter. They weren't, and at 18-24V differential supply it just overheated like crazy, but never melted and worked fine afterwards.

One reason for high ratings is they may have short-circuit protection, and you can heatsink DIPs if you want to push them.

>> No.1607121
File: 30 KB, 503x261, editor_201706081496908851942302.jpg [View same] [iqdb] [saucenao] [google]
1607121

>>1607090
Anyone? Is this connection legit?
(ignore ground obv, since I'm lazy)

>> No.1607123

>>1607121
for sure!

>> No.1607133
File: 227 KB, 600x780, perf.png [View same] [iqdb] [saucenao] [google]
1607133

>did my first perf board layout
Wew, that's gonna be tight.
Guess I better practice on some simpler circuits before soldering this together. Then again, how hard could be it?

>> No.1607140

>>1607133
Circuit diagram missing.

>> No.1607147
File: 328 KB, 2155x803, schematic.png [View same] [iqdb] [saucenao] [google]
1607147

>>1607140
not in the schematic but included in the perf board layout is a mod for switching between clipping diodes (top left) and a mod for adding in 2 different sets of resistors/caps in parallel with R7/C6 for a bass boost toggle (bottom left)

fingers crossed that I didn't fuck up

>> No.1607149

>>1607108
Measure open loop voltage and short circuit current. No need to speculate.

>> No.1607152
File: 1.66 MB, 1743x805, components.png [View same] [iqdb] [saucenao] [google]
1607152

I'm new.. How important are the specifications for components?

On some shopping lists I just see really simple terms listed, but nothing about voltage, tolerance, etc etc etc. The example I'm working off of is pic related.

If I get that resistance and shape, is that all I need or would I be potentially buying parts I don't need? basically, how important are the individual specs

>> No.1607157

>>1607147
Thank you. Looks like stomptronics. Why the JFET instead of a BJT emitter follower?

>> No.1607159

>>1607157
Dunno, that's what's in the original circuit so that's what I'm going with.
There's probably a good explanation to be found here: https://www.electrosmash.com/proco-rat

>> No.1607160

I want to get started on some Arduino projects, but I have no idea how to code. Is there any way to go around it or should I just start learning how to code?

>> No.1607161

>>1607152
resistors have a resitance and a power, if you exceed the power rating they will burn and catch fire. its hard to exceed the typical power rating (0.25W) (actually the small ones are probably less, 0.125W) unless you are trying to drop a large voltage for an led or something mad.
tolerance is typically 5%? 10%? i can't even remember because it doesn't matter unless you need something precise you will know all about it.
capacitor has a voltage rating, sometimes they are really low and might fuck you actually but 3v or something circuit (2xAA battery) probably can't go wrong.
ics have lots of ratings for voltage, current per pin etc.

>> No.1607169
File: 35 KB, 406x500, 1542223143083.jpg [View same] [iqdb] [saucenao] [google]
1607169

>>1606907
with gcc on the bare metal, a blinky is possible in 90 bytes
>30 bytes vector table
>2 bytes for the bad interrupt halt
>4 bytes for _exit to clear interrupt enables and halt
>12 bytes to set up stack pointer and get to main()
>42 bytes for main() itself

>>1607133
I don't like those horizontal mount 1/4W resistors with only 0.300" between leads. you have to make the bend very close to the body, which is a lot to ask without a bending die. of course if they're only 1/8W carry on
generally, start in the middle of the board, use a 1.2mm screwdriver tip, and leave the offboard connections for last. good night and good luck

>>1607152
>How important
very. look at R9 and R10 for ex. there are many different configurations of those little yellow epoxy-dipped SIPs, some with just five separate resistors inside, some with 9 resistors tied on one side to a common terminal, some with 2 banks of 8 resistors each bank tied on one side to a common terminal forming 8 voltage dividers, and so on. on the other hand, for most circuits, resistor-to-resistor matching inside the network doesn't really matter that much
anyway for hobby purposes, it's conventional to choose 1/4W ±5% resistors, ±20% electrolytic capacitors with any voltage greater than the circuit's power supply, 50V (or greater) ±10% or ±5% ceramic capacitors unless otherwise specified. ICs are all different and subtle, and substitution is fraught with loads of traps for young players
>buying parts you don't need
this is how you build up a stock of components for your hobby, only degenerate when taken to an extreme. seriously don't get suckered into paying $0.10 for one resistor when you can get ten for $0.12

>>1607160
that's the whole point of arduino: to code
you might check out the arduino general

>> No.1607175

Is Rated Power important for a rotary potentiometer?
The part I'm trying to replace says "Power (Watts): 0.5W, 1/2W" and I have a part that says Power Rating: 0.05W

>> No.1607177

>>1607175
>I have a part that says Power Rating: 0.05W

That might be a typo, or a mighty small pot. Whatever the actual power rating is, it's as important as any power rating, in that you want to stay comfortably below it unless you know your environment will keep the temperature under control.

>> No.1607180

>>1606147
"Working" in the basement. Uh-huh.

>> No.1607182

>>1607161
no point buying 5% resistors now that 1% are practically the same price unless you need thousands.

>> No.1607188

Broad question but:
Are Chinese clone chips for certain old ICs problematic? Like an old texas instrument chip that's around 5 times cheaper than from the actual manufacturer:
https://lcsc.com/product-detail/Others_Texas-Instruments-TI-SN74LS245N_C352966.html
but on Ali it's around a quarter of that.

>> No.1607189

>>1607188
Pretty sure all electronics are chink shit general these days.

>> No.1607191

>>1607188
for certain old chips where differences are harder to measure, it's a risk e.g. LM358s have been seen re-marked as NE5532s. re-marking has been seen even where the difference is obvious e.g. some random power supply controllers have been seen re-marked as ATmega8s
that said, the '245 could be new-old-stock, could be a clone, could be pulls, could be a substitute from NXP or whatever. leaded parts will tend to trade at a great discount on the spot market as the contract market for them declines
not that there's anything morally wrong with pulling 74 series chips. 25 years ago I used to visit a junkyard on the coast with a bug rug and a chip puller and bought 'em by the pound

>> No.1607215
File: 221 KB, 597x784, SMALLER.png [View same] [iqdb] [saucenao] [google]
1607215

>>1607133
WE MUST GO SMALLER

>> No.1607217
File: 12 KB, 400x400, 1497301724457.jpg [View same] [iqdb] [saucenao] [google]
1607217

>>1607215
I've been playing with this for like 8 hours now. It's really just a fucking addictive puzzle.

>> No.1607228

>>1607217
isn't it, though? I predict a smooth leveling-up for you when you have etched boards, multiple layers, vias, and so on
I suggest just two things you would do well to develop now: a sense of diminishing returns (when have I fucked around with this layout enough?), and an appreciation of design for manufacturability (can I build this? how much will I hate myself if I need to build another one? what if I need to replace R9 or R10? have I left room enough for enclosure protrusions or fasteners?)

>> No.1607229
File: 38 KB, 600x450, 1452361597116850892.jpg [View same] [iqdb] [saucenao] [google]
1607229

>>1607215
Go 3D.
I always hated PCBs. When I got a microscope, first thing I did was to solder an NE555 space-mount blinker. I lost photo, but I really liked it.

>> No.1607234

Okay, I wanted to bring unsold powerbrick to recycler, but >>1606456 happened.
You DIY you lose.

And yeah, pumping 2amps at 41V (voltage on battery) for 10S battery from china with unknown BMS is sketchy AF...

>> No.1607237
File: 210 KB, 2736x624, 3109404287625376496.jpg [View same] [iqdb] [saucenao] [google]
1607237

>>1607234
fucking attachment. FUCK

>> No.1607239

>>1607229
>not making PCBs as narrow as the DIP itself and wrapping the pins under it
The real brainlet filter is constraining yourself to how the footprints were meant to be used.

>> No.1607243

>>1607239
This wasn't a DIP, but rather SOIC-8 NE555 in my case. With 3D you can make shit tiny. It would be impossible to achieve on PCB

>> No.1607247

>>1607243
But how do you use something with no mounting points? Half the advantage of a PCB is the holes by which to screw it into a case. In my case I was designing an FM radio to fit inside a pen, with the DIP MCU itself being wrapped around the long-thin PCB. With your 555 thing I can't imagine how you'd connect it to anything, aside from just soldering wires. A small PCB also means your wires aren't prone to vibration or work hardening.
In the case of that "one chip arduino" it makes some sense because you still have the intact DIP pins to insert into a socket, but it's still a bit clumsy since you can't actually solder it into a project without potentially reflowing the connections already on those pins.

>> No.1607253

>>1607247
>But how do you use something with no mounting points?
Hot sperm gun, cast in resin.
>Half the advantage of a PCB is the holes by which to screw it into a case.
Nah. PCB allows automation and ease of troubleshooting.
> With your 555 thing I can't imagine how you'd connect it to anything, aside from just soldering wires.
Some components can be soldered on pin, some will require jumpers.
Good 3D-assembly should require minimal amount of jumpers.

>> No.1607263

Probably a decent place to ask, I live in a remote and rural area, due to personal reasons (I'm shit fuck crazy paranoid) I want to secure my 30 acres against Intruders. When I bought the place the front door had the window smashed. Kids they say. I need a way to record and watch as far as 1\2 a mile from my home reliably without using cell towers. WiFi is out the door, unless they make 800 mhz cameras. Is their any other way to transmit this far? Some other non used frequency that's good for short range and won't get me in trouble? Building a perimeter fence. Nothing fancy, just 8 feet tall 2"X2" wire fence. I want to run a real nasty bull electric fence. Something that will hurt. Maybe some sort of voltage sensing circuit so if it's cut I know.

>> No.1607282

>>1607263
Build your fence using a car battery, a turn signal relay (also from car), and an ignition coil. Put a light at the end of the circuit to signal when it's working/not.

>> No.1607288

>>1607263
I would connect live wire through a capacitor (so current is limited to 5 mA).

>> No.1607291

>>1607263
I am assuming that the cameras and transmitters will be powered from a solar/battery combo, otherwise the correct answer to "how do I power the cameras" is the same as to "how do I shot data"
>wifi
why is it out the door? wifi transmitters' legal output range is (in burgerland) regulatorily limited to 1W total output power. whether you sprinkle that in a nice round circle around you or shoot a tight beam to and fro is up to you. directional antennas can be built easily inside a pringles can and/or a PVC pipe with some threaded rod, some nuts, and some washers
once you have read this informative FCC publication on its "Part 15" (unlicensed low-power devices) rules, please pass it along to our electronics enthusiast friend in Ohio https://transition.fcc.gov/Bureaus/Engineering_Technology/Documents/bulletins/oet63/oet63rev.pdf but tbf I can't see them honestly giving a fuck what sort of microwave oven you're living in on your own 30 acres, as long as it's not leaking
don't most fence chargers have terminals going in both directions? even if not the other wire's got to end somewhere. a neon lamp with a hueg resistor to ground should glow when the far end is connected and not when it isn't, without draining too much oomph off the fence wire

>> No.1607333

>>1607282
yep, it's working
nope it's not
yep
nope
yep
nope
yep

>> No.1607337
File: 43 KB, 443x329, LM331_FVC.png [View same] [iqdb] [saucenao] [google]
1607337

I need to make a frequency to voltage converter (FVC), I've found this circuit that utilizes an LM331 and one of its configurations (the FVC one), but I want to know if I can supply this circuit with +5V or if any of you know an alternative to this IC or any other circuit to perform this task.

This is the datasheet: http://www.ti.com/lit/ds/symlink/lm331.pdf

Pic related is the schematic.

Sorry for the bad english.

>> No.1607338

>>1607337
>This is the datasheet
Now read it.

>> No.1607340
File: 27 KB, 640x640, TANTCAP.jpg [View same] [iqdb] [saucenao] [google]
1607340

https://www.digikey.com/products/en?keywords=478-8961-1-ND

I can't find this part for a reasonable price.Am I just shit out of luck? Using it for an AdLib clone

>> No.1607351

>>1607340
use a couple of ceramics in parallel?

>> No.1607352
File: 131 KB, 301x199, 106.png [View same] [iqdb] [saucenao] [google]
1607352

>>1607340
Don't forget the 476 and the 107.

>> No.1607356

>>1607352
That would actually work in place of it?
a TAP106K035SCS I assume

>> No.1607362

>>1607288
You will never be able to kill an adult human with just 5ma

>> No.1607369

>>1607333
And when it stops flashing altogether?

>> No.1607390

>>1607362
Oh, you need to kill?
Than voltage doubler with giant ass caps on the output

>> No.1607392

So if there's two capacitors and they're both identical but one is 100V and another is 50V, and the BOM doesn't specify a particular Voltage Rating... Do I just pick the 100V? Is it just plain better or something?

Sorry, am very new.

>> No.1607396

>>1607369
definitely don't work
and by then your battery's probably fucked too

>>1607392
the 100V cap may be less attractive in other ways, e.g. a wider capacitance tolerance, a larger voltage and/or temperature effect on capacitance, physically larger, cost

>> No.1607398
File: 15 KB, 170x300, Singchair.jpg [View same] [iqdb] [saucenao] [google]
1607398

>>1607390

>> No.1607411
File: 340 KB, 1536x2048, IMG_20190509_134829.jpg [View same] [iqdb] [saucenao] [google]
1607411

Yes! My beautiful tiny babies are finally here all the way from china
Time to upload some leet hax code

>> No.1607416

>>1607411
what kinda chups you got there bud

>> No.1607417

>>1607411
tiny attiny's 85

>> No.1607418

>>1607396
Also ESR, life/temperature rating, maybe more dodgy manufacturer.

>> No.1607422

I have a list of resistors and capacitors, but the bill of materials doesn't state anything over than the resistance and capacitance, respectively. No voltage rating, watts, tolerance, etc...

Is it even important? Can I just plug in "Carbon 15K Ohms" etc and buy the cheapest results and be OK? I thought the specs had to be very specific.

>> No.1607429

>>1607398
I said really large caps.

>> No.1607434
File: 205 KB, 633x807, EVEN SMALLER.png [View same] [iqdb] [saucenao] [google]
1607434

>>1607215
Bet you could fit this in a 1590A box if you really tried. Although there'd probably not be enough space for 4 pots, a DPDT toggle switch, a 3DPT stompswitch, 2 1/4" jacks, a power jack and a status led.

>> No.1607451
File: 57 KB, 600x470, Clipboard01.jpg [View same] [iqdb] [saucenao] [google]
1607451

Hmm, i got this programmer with the attiny chips, thinking it will save me some hassle to be able to program them via usb, but at close inspection i don't think i can.

the attinys don't have any usb interface because the are 2 tiny for that and this dev board has no chips on it apart from som caps voltage regulators and such

Fuck looks like i will have to use UNO as a programmer , but at least i have nice little breakout for the tinys

>> No.1607453

>>1607451
The USB port is probably just for power.

It's programmed over that unsoldered header.

>> No.1607458

>>1607337
By the way, you can make a frequency to voltage converter with a capacitor and a resistor.

>> No.1607462

First time posting here so bare with me and ask for clarification on anything I'm not clear enough about.

I'm currently at the start of a refit on a boat. The boat is destined to spend it's time in cold areas cruising around the arctic circle, Norway, Canada, Greenland etc.
We want electric propulsion for the fact it has better torque and is better adjusted to maintain RPM on the propeller. BUT batteries hate the clod and a conventional engine is a valuable source of heat for the boat.
I want to do a electric hybrid setup with a diesel or gasoline engine running an electric motor. Except i'm unsure on what rating generator I need compared to the motor. It seems HP and KWh are the units of business but a 20KW generator and a 20KW electric motor seem to be disproportionately sized.

TL:DR
>Electric motor powered by conventional Generator
>What does a dumb dumb need to know before starting
>Is the KW rating of HP rating what I need to focus on

>> No.1607465
File: 393 KB, 2048x1536, IMG_20190509_154045.jpg [View same] [iqdb] [saucenao] [google]
1607465

>>1607453
>unsoldered
Not anymore.

Don't my my humble ts100 flex.

>> No.1607467

>>1607465
*mind

>> No.1607481

>>1607462
HP and kWh are units of power. Both of em. An electric motor will have an output power, probably measured in horsepower with efficiency accounted for if it's a Burger engine, otherwise just use some random conversion like https://sciencing.com/calculate-footpounds-energy-6002311.html

Btw you should use a suitable battery in-between the generator and motor, because otherwise the power would be really unstable and wear hard.

>> No.1607485
File: 3.73 MB, 2976x3968, IMG_20190509_103129.jpg [View same] [iqdb] [saucenao] [google]
1607485

Hey /diy/

I have this toy that I am hoping to add a 3.5mm jack to so that I can hook it up to an external speaker.

I know I will probably need some sort of amp (I'm looking at those mini mp3 player amps) before I connect it to the external speaker, or it still won't be very loud.

So my question is this, can I just solder in a jack via the point where the speaker is currently connected?

Pic related.

>> No.1607518

>>1607481
Yea, I'll need a battery bank to power the navigation gear, radar vhf etc.
I'm trying to keep it as basic as possible so Lead something batteries will be best. (weight is also good for ballast).

Thanks for the link too!!

>> No.1607524

Complete beginner here, I've just bought the books listed in the Principles section as well as this kit that's supposedly for the Make: Electronics book. Anyone bought this before and is willingly to share their experiences? https://www.amazon.com/gp/aw/d/B07LDJ2Y7M/ref=ox_sc_act_image_1?smid=A14UHVV7U39CHV&psc=1

>> No.1607533

should I get the cheapest soldering iron out there

>> No.1607535

>>1607533
Depends. Some are okay, some are garbage.

>> No.1607545

>>1607533

a cheap iron can be a major source of frustration that can put you off of electronics forever. but this is a good thing; there's way too many damn noobs with a million noob questions in this hobby.

>>1607524
> I've just bought the books listed in the Principles section

who buys books in 2019!?

as for the kits, they're okay for getting stuff that works, but they're very poor when it comes to fundamentals. still, i dont know of anything you can buy that covers both theory and practice in a comprehensive way. (maybe you can invent such a thing and become a millionaire.)

>>1607485

yep, just replace the speaker with a jack. in fact, if you research the topic, there's a way to have the internal speaker go silent when you plug in the external one. one way to avoid using an amplifier is to use a very efficient external speaker.

>>1607458
> you can make a frequency to voltage converter with a capacitor and a resistor.

but the result is guaranteed to disappoint.

>> No.1607575

>>1607545

Cool, thanks good anon!

>> No.1607602
File: 31 KB, 632x412, Unbenannt.png [View same] [iqdb] [saucenao] [google]
1607602

What could go wrong if I use a transistor instead of a mosfet for overdischarge protection?

>> No.1607620

I'm trying to implement this analog front end with the op-amps included in the PIC32MK1024GPD064-I/PT. http://ww1.microchip.com/downloads/en/DeviceDoc/PIC32MK-General-Purpose-and-Motor-Control-%20(GPMC)-Family-Datasheet-60001402E.pdf#M37.9.64026.TableTitleDS.TABLE.121.Comparator.Specifications gives the minimum stable closed-loop gain, OA8 and OA16, as 8. If I'm understanding correctly, the inverting amp is fine as this only applies to non-inverting configurations. Is that right? What about the sallen-key filter? I guess I could just implement that with one of the unity-gain stable buffers that are part of the same peripheral, but I'm curious if it'd even be a problem. I haven't really had to deal with op-amp instability before so I'd love to take this opportunity to learn.

>> No.1607621
File: 28 KB, 749x335, file.png [View same] [iqdb] [saucenao] [google]
1607621

>>1607620
Forgot the pic.

>> No.1607623

>>1607602
Tranny current is dependent on base current if i am not mistaken

>> No.1607624

>>1607133
>>1607215
>>1607434
Jesus. Just give China your $20. You'll save enough time for it to be worth it. And if you haven't done layout before this is a simple opportunity to learn a valuable skill.

>> No.1607628

>>1607624
Having a PCB fabbed costs 20 bucks? Lmao, then I'm definitely not doing that. That would double the cost of a pedal. Especially for the one offs I'll be doing at first.
Maybe I'll get some FeCl3 and etch my own though. Which I might get anyway for etching the aluminium enclosures.

>> No.1607629

>>1607465
What soldering iron guard is that. Unironically I think I have that iron. and a 3d printer.

>> No.1607631

>>1607628
Having ten fabbed costs $20. Do you not live in the US or something? Anyone posting on an electronics forum should be making $20 in well under an hour.

>> No.1607634

>>1607602

wouldnt work coz the chip uses the FET's impedance to measure over-current. a BJT would alter that characteristic majorly.

>> No.1607635

>>1607631
Nope, Netherlands. But it's not really about that. I'm just a penny pinching asshole. Why take the easy way out when you can save money and do it the hard way?

>> No.1607641

>>1606134
I'm looking for a rectangular connector. I want to replace a propietary, rare and expensive connector on some automotive electronics with something widely used (internationally available), cheap, using altogether 91+ pins (it's like two sockets joined as one piece, with one having 13 pins and the other having a set of 64 next to a thicker set of 4).

I've had a look at a few options, if there's a standard, non proprietary automotive connector I'd be super happy, failing that I've tried having a look for what a suitable standard socket might be from aircraft or something.

A few more pins wouldn't be bad and may be pretty useful btw, everything I've looked at tho is fucking expensive, like it'd be way cheaper to replace the whole unit a few times over.

>> No.1607642

>>1607631
>why don't you just pay someone to do <x>?
>/diy/
How about you get out?

>> No.1607645

>>1607629
>I think I have that iron
>I think
you would be pretty sure if you had the best soldering iron on the current market

anyway it's just some guard i found when i typed ts100 into search on thingyverse

>> No.1607646

>>1607641
scart connectors? IDE has something like 39 pins

>> No.1607647

>>1607645
It's not my iron, and I'm too lazy to check. Made it easier to describe for anonposting. Don't develop autism for soldering irons please.

>> No.1607649

>>1607642
You always buy something at some point unless you're digging your own materials out of the earth. It's just a matter of what point in the chain you're buying stuff at. If you buy PCB fabs you can save time on soldering and troubleshooting and spend that time on designing. Plus you can use SMT components which are worlds better, and you have a skill you can show to employers that's far more valuable than dragging wires around on a protoboard website.

>> No.1607653

>>1607646
The look a bit like scart connectors, but the pin density is higher. I need to get a pic of them maybe with a ruler, but I'll have to do that tomorrow.

>> No.1607656

>>1607641
https://en.wikipedia.org/wiki/SCSI_connector

>> No.1607665

>>1607631
>>1607649
So imo there's two sides to this, but if you're looking at most bog-standard working for bossman electronics jobs there's no real point in experience of fabbing your own boards, just design them and get them out. Sure, that's true, no worries, and a fair number of people in that vein are aggressively against home fabbing because they're like that.

For people outside of that, so working for themselves or for people that aren't... overly particular about prototyping, knowing how to fab a board can mean you get to be hands on with a design within hours, with a Chinese fab you're getting probably a better margin but you didn't need 5-20 copies yet and getting them took weeks.

So not a bad skill, but not an appreciated skill.

>> No.1607667

>>1607656
Didn't think of that, promising, thanks.

>> No.1607710

I've got an HP 54501A that will not turn on. Obviously the power supply is shot. In the reverse engineering document (http://www.radiocollection.be/images/restaurations_img/HP54503/supplyHP54503.pdf)), the resistors R13 and R14 are blown. Reading the resistors, they seem to be 8 ohm resistors. I replaced them with two new ones, and they promptly asplode. Clearly something is drawing way too much power. Does anyone have any ideas for how to move forward? Would the transformers or electrolytic caps be suspect? Or does anyone have experience repairing this?

Additionally, what's the smallest Marine Deep Cycle battery you guys know about? I need ~100 CCA.

>> No.1607739
File: 29 KB, 426x281, Capture.png [View same] [iqdb] [saucenao] [google]
1607739

>>1606134
Hey guys, I'm new to electronics and I've been looking at this circuit and I wanted to test my understanding.

From what I see:

-positive charge flows in at Vcc and at first most of the current flows toward charging the 470uF capacitor

-eventually the capacitor is fully charged and it behaves like an open circuit

-around this point the threshold is passed, causing a high out at pin 3 and opening the discharge pin to ground, causing the capacitor to discharge through pin 7

-the positive voltage at pins 2, 7 and 6 drops until it is below the pin 2 trigger, closing 7 and resetting the circuit to charge 7

If I got anything wrong any corrections would be much appreciated

>> No.1607741

>>1607422
the circuit will still work if you build it with $5 audiophile capacitors or 5¢ ceramics. if they don't spec tolerances etc., they probably don't matter much in this case and jellybeans* will work fine
as for cap voltages, look up Kirchhoff's voltage law and see that no cap in the circuit will see more than the highest potential in the circuit so just round up from the supply voltage when you order
*components that are cheap, undistinguished, easy to find, and bought in bulk

>>1607451
>>1607453
look up USBasp, noobs

>>1607649
a beginner can still gain a useful sense of topology and other aspects of the art of pc board design by designing to a fixed grid

>>1607710
could be Q3 blown, could be a shorted turn. switching power supply repair is not for the noob, so if you are in doubt, refer to a qualified person

>> No.1607742

>>1607417
Oh man, I ordered a couple of those a while back, but I figured they were a bit too pricy for general use so ordered some 13s instead.

>>1607451
These can be programmed with the micronucleus code through the arduino environment, just follow a digispark tutorial.

>> No.1607751

>>1606369
You should be able to get a uC that'll wake up from low power mode when it sees an edge on a pin. I'm pretty sure there's PIC18s that'll do that in a DIP-8 package. Otherwise, there's definitely PIC24s that do it. All you gotta do is have it wake on a loss of continuity, and do a little debouncing just to be safe.

>> No.1607757

>>1607237
>>1607234
Okay.
41.1V... 1.6A
How do I put down lithium batteries, in case Chinese batteries don't have overcharge protection?

>> No.1607758

>>1607741
It does appear Q3 is blown. Now I get to try and track down a replacement for a TIPL755A.

>> No.1607763

>>1607742
pro tip: AVRs are often cheaper on Digi-Key than on alibay

>>1607751
I can't think of any micro with a sleep function but without wake-on-pin. it's the first and most obvious answer to "now that we can put the core to sleep, how do we wake it up?"

>> No.1607802

>>1607751
>wake up from low power mode when it sees an edge on a pin
Isn't that the external interrupt I was talking about?

>>1607763
>AVRs are often cheaper on Digi-Key than on alibay
I'll keep that in mind, but since I never order stuff from there I'm never close to their "free shipping limit", which is especially problematic since I'm not exactly in North America or Europe. Last I checked Arrow had one of the lower "free shipping limit"s, and I have considered buying some more specialist parts from them in the past for that reason, but as I'm currently pouring a significant amount of my cash into this new/old bike, that will have to wait.

>> No.1607808
File: 8 KB, 400x400, tegaki.png [View same] [iqdb] [saucenao] [google]
1607808

>>1607741
>look up Kirchhoff's voltage law and see that no cap in the circuit will see more than the highest potential in the circuit
>he doesn't know

>> No.1607815

>>1607758
if it fails to come back to life after you re-replace the coil and the transistor, the transformer is probably shot
without a scope, full specs of the transformer, and probably a variac, that's about as much as you can do

>>1607808
often, great truths are taught by supposing little half-lies, dispatched one by one as they become no longer necessary for understanding. pedalfags and synthfags can often get through life without ever seeing an inductor

>> No.1607821

>>1607458
I don't think the results would be good

>> No.1607858

>>1607815
Perhaps. Also I thought of a method for stepping up voltage that doesn't require any inductor or switching or AC:
>get a bunch of LEDs in parallel
>get a bunch of photodiodes in series
>put them both in a reflective chamber
>drive the LEDs and get higher voltage out the other end
Could probably do the same thing with peltier devices or possibly other transducers too. Though I'm not sure which would be less staggeringly inefficient.
>inb4 light is THz AC

>> No.1607885

>>1607742
>13s
I looked at those but they were missing some vital features i needed like i2c plus 85s have more memory and unless you are mass producing the price difference is pretty small

>> No.1607886

>>1607885
>but they were missing some vital features i needed like i2c
bitbanging i2c is easy

>> No.1607893

>>1607886
With only 1KB of program memory? Perhaps not.

>> No.1607896

>>1607893
pretty much, 1kb of memory is fine if you want to make a button to turn a lamp on or off, but for anything even slightly more complex where you need more code and a library or two you are fucked.
8KB is pretty much idea amount of memory to have on hand
And like i said, if you don't mass produce then a price difference bewteen 20 85s and 20 13s is negligible

>> No.1607897

>>1607893
You only need a little more memory to bitbang it then to set up hardware i2c.

>> No.1607898

>>1607742
>digispark
That only works if you have the digistump attiny which has an usb com chip on it
Which i don't know why would you ever use, since it has stuff like leds and voltage regulators that waste power, and i mean will you hot glue a digistump board with usb connector sticking out of it, that needlessly consumes power with power leds and such into your project, even though you don't need any of those parts atfer you program it?

With normal attiny bug you need some programmer external programmer like uno that can interface with usb

>> No.1607899

>>1607896
I mainly bought them for projects where I just need to control something I couldn't easily control without an MCU, like an RDA5807, IR remote control, or something with rotary encoders.

>>1607897
Yeah, I thought it was unlikely that I2C in the 85s isn't programmed in some sort of ROM.

>>1607898
>That only works if you have the digistump attiny which has an usb com chip on it
No it doesn't, there's no such chip on it at all. It just has the 5V linear regulator and the ATtiny itself, and a bunch of passives. Check the documentation (here's the schematic: >>1606416), the whole point of Micronucleus is to allow programming directly into the pins of the MCU. I have one in front of me for the purpose of prototyping designs that I'd later copy straight onto a DIP ATTiny.

>> No.1607902

>>1607899
>I mainly bought them for projects where I just need to control something I couldn't easily control without an MCU, like an RDA5807, IR remote control, or something with rotary encoders.
Should have bought the 3cent MCU meme instead.

>> No.1607903

>>1607898
>>1607899
Oh and I have one of the micro-USB Alibay boards like >>1607451 for the purpose of programming the things, and as you can see it has the same zener diodes and other passives on the USB data pins. I could count and compare the number of passives, and I will:
Digispark:
>3 diodes
>5 resistors
>2 capacitors
>2 LEDs
>2 ICs
Chinky USB board:
>3 diodes
>5 resistors
>3 capacitors
>2 LEDs
>2 ICs
I wonder what that extra cap is for, possibly input filtration for the linreg.
I'd be using it right now if my most recent os update didn't fuck with the >arduino platform, so now I'm trying to figure out how to get >debian onto an SD card so I can use that instead.
Did you think all the (13) passives in that guy's pic were just for show?

>>1607902
>one time programmable
>no ez usb
Maybe when I'm a little more competent.

>> No.1607920

>>1607075
Wow, great answer to my ambiguous question! Also within what I assumed but did not know.

Do you have a source for this info or by chance experience working in semi industry? I'm most interested in working for one when I graduate but feel like they'd be super selective, which I do not have the gpa for that

>> No.1607921
File: 270 KB, 2048x1536, IMG_20190510_133504.jpg [View same] [iqdb] [saucenao] [google]
1607921

Mother of god, it's working!

No more shitty ironing of torn laser printed magazine papers on pcbs, no more shitty gaps coloring and hunting all the imperfections.

Beautiful flawless industrial grade pcbs, here i come

>> No.1607922

The AVR on my generator died. Can I just buy the component parts and rig one myself or is there some sort of chip or processor that I would have to code?

Also does any AVR of equal ratings work with any generator? I don't care about things looking nice and fitting perfectly back together.

>> No.1607923

>>1607921
>that pic
>no more shitty gaps coloring and hunting all the imperfections
>Beautiful flawless industrial grade pcbs
this is true comedy

>> No.1607925

>>1607923
The paper wasn't properly attached and the line settings like width and shit aren't properly tuned yet.
On perfectly stationary, noncrumpling pcbs with perfectly tuned sharpie line width, the results will be nearly flawless.
I am getting a semichub just thinking about it

>> No.1607930

>>1607925
Well I wish you luck in any case. Personally I print onto transparencies for PCBs, but for those with a 3D printer who don't want to buy $15 of photoresist and carbonate and such I guess your method is acceptable. I'd also consider a much finer tip marker, like a 0.5mm permanent.

>> No.1607931

>>1607930
The great thing is that i can also replace the sharpie with a small motor and directly cut the pcb instead AND drill any holes it needs as well
so no nasty acid handling
And maybe i will feel adventurous one day, i will get a one of those small vacuum smd pick up tools, attach it to the printer and turn it into a pick & place machine as well as pcb cutting machine

>> No.1607932

>>1607931
>holes
Yes that's definitely an advantage. At my maker space I'm considering using a vinyl cutter to use stickers as etch-resist, and I could arguably use the CNC router for drilling my holes provided I can get my tiny PCB drill bits to work in their collets.
>3D printer pick & place
Now that would be a sight to see.

>> No.1607934

>>1607932
>Now that would be a sight to see.
Several people already made it
https://www.youtube.com/watch?v=EuLc1wDoxgI

Difficults depends on features you want, like do you want it to also apply the solder paste? do you want it to be able to rotate parts it picks up? Should i be bale to peel back the tape from the smd spools?
But i won't really be making it any time soon, since for the small volume of hobby project pcbs i output it would be meaningless for now

>> No.1607954

>>1607921
make it draw a penis

>> No.1607964

what is faster to learn if you only want to make simple pcbs with simple components, kicad or eagle?

>> No.1607969

>>1607964
kidcad has better crayons.

>> No.1607972

>>1607954
nice try, /lgbt/

>> No.1607973

>>1607903
>how to get debian onto an SD card
As a live OS? dd/win32diskimager/rufus whatever

As a real OS? install it using ext2 as the filesystem.

An important thing is normally you can't boot from an SD card slot - they're either not mounted properly, or they just aren't entered as available boot devices. If you're using a USB card reader, you should be able to do it, but you may need to install in a non-UEFI way (if it doesn't boot in UEFI mode, try first), which could mean enabling legacy support in bios.

>> No.1607987
File: 46 KB, 550x547, 1539739001478.jpg [View same] [iqdb] [saucenao] [google]
1607987

Okay I know this is a retarded question but I don't know jack about electronics so bear with me.
After reading about Ohm's law I was left with the impression that the voltage is always equal to the resistance times the current. However, I own several phone chargers; one is rated at 5V/1.5A, another is rated at 5V/2.1A and and another is rated at 5V/1A. I can use the same USB cable to connect each charger to my phone. Since the voltage and resistance of the cable remain constant, how can the amperage change? Do smartphones have some built-in dynamic resistor that can adjust the resistance to match the rating of the charger?? If that's the case, does it have a technical name I can look up to find out how it works

>> No.1607994

>>1607987
The load draws only what it needs.

If you have a 10 ohm load and 5V power source then current will be 5/10 = 0.5A.
How much current the source can source is meaningless, in fact the PSU in your computer can output something like 20A on the 5V red rail.
You can hook this up directly to your phone and it will charge just fine without any damage.

>> No.1607996

>>1607987
>If that's the case, does it have a technical name I can look up to find out how it works

Just keep studying electronics. Ohm's law applies to a simple DC voltage across a resistor. Battery chargers tend to be complex and sophisticated circuits.

Your question is sorta like saying you just learned that uranium is an element, so why does it explode in a bomb and doesn't blow the earth up while it's in the ground.

>> No.1607997
File: 24 KB, 490x317, unscrew.png [View same] [iqdb] [saucenao] [google]
1607997

Just the facts, ma'am, just the facts.

>> No.1608011

>>1607964
Wow this i great.
I am trying kicad and watching a tutorial that is barely a year old the the interface is COMPLETELY different, even has different names for everything.
This makes it real fucking fun to follow the tutorial

>> No.1608015

>>1607858
Interesting. Depending on how many you could fit in a package it might even be useful since you don't need any external components.

>> No.1608032

>>1607987
Think of a pipe with water being pumped in. The diameter of the pipe would be the resistance. Bigger diameter, lower resistance (more water per cross section of the pipe). The pump puts in water at a certain force (voltage), which makes it flow (current). Now imagine the pipe being closed off at the other end with a faucet with a maximum opening equal to the width of the pipe. The load is connected to the faucet and will open the faucet to get the amount of current it needs. It will only get what it needs, but it can never get more flow (current) from the pipe than what the width and pressure (resistance and voltage) dictate.

If this is completely retarded please correct me, but I really like water analogies for getting an intuition for electronics.

>> No.1608064

how do I learn how to make RF circuits? I tried searching up "FM receiver diagram" but it was too complex for me. Where the fuck do I start? I read the wikipedia article and apparently it's just filtering the signal with a resonant circuit with the right capacitance/inductance, amplifying the signal, demodulating it and amplifying it again. But when I look at a diagram I don't understand what's what.

>> No.1608068

>>1607858
Actually, I remembered that something similar exists. Piezo transformers.

>> No.1608069

>>1608032
while I wouldn't say you're wrong a. you didn't specify the analogous items in your scenario so b. based on your argument when you said
>never get more flow (current) from the pipe than what the width and pressure (resistance and voltage) dictate

which suggests
>the pipe connecting to the faucet (the wire connecting to the load) is the determining factor for how much current the load consumes
which is not true, in this case.
the faucet is the key element in this case.

>>1607987
so yeah if this is a non-technical question, they do have a built-in dynamic resistor of sorts, like a MOSFET (your name) to control ('request') the current. unlike the voltage, if your supply listed 500mA it will charge much slower, and if it was not limited may overheat the transformer inside.

but the MOSFET generalization is just that, depending on the battery voltage (regardless I'd imagine) there's probably a switchmode power supply in your fone that takes a voltage higher/lower and outputs what it needs to charge the battery

>> No.1608073
File: 130 KB, 836x655, Fm-radio-Receiver-circuit-diagram-1.jpg [View same] [iqdb] [saucenao] [google]
1608073

>>1608064
>FM receiver diagram
if you referring to the first ckt that pops up on google (using lm386) you've got some work to put in bud. I haven't really worked with RF but you're in the right ballpark. Regardless explaining it simply is much easier than building it. if you don't know how pic related works then you need to understand transistor operation which can be done by reading art of electronics (you can pirate). It might even have stuff on rf

>> No.1608097

>>1608073
That's not a typical FM receiver and it is barely usable. You will not understand its function until you discover that it's also an oscillator that inadvertently transmits at the same frequency it it supposed to receive and make audible. That thing is missing everything you expect from a decent FM radio, e.g. sensitivity, selectivity and AM suppression.

>> No.1608190

>>1608011
KiCAD 5 was a pretty major upgrade, and a welcome one. some of the

>>1607972
it's not gay if it's Dickbutt

>>1608064
arrl.org has some books and plenty of articles. you might even be able to pir8 some
the licensing study materials for amateur radio in general are pretty good. by the time you can pass the highest level exam (called Amateur Extra class in FCCland) you'll have a very good idea how to build a radio receiver or transmitter from scratch (if your component stock is suitably endowed)
>FM receiver
can be subtle, as you've seen with that 2-transistor thing. the more complex ones leave less to chance and might be easier to understand. fwiw most broadcast FM receivers are one-chip devices with all the active functional blocks on the chip, and all the designer needs to provide are filters and bias

>> No.1608193

>>1608011
>>1608190
(continuing)
library management functions were sorely in need of rethinking from 4, as you have probably seen. the symbol librarian was way different from the footprint librarian and a bit awkward to use. it's quite a bit more rational now

>> No.1608221

>>1607973
Live image. I gave up and slapped it on my USB 2.0 drive since apparently that's fast enough for simple stuff, not to mention more ubiquitous. I was trying to use something called unetbootin but it wouldn't recognise the SD card regardless of how I reformatted it. If I care enough in the future I'll buy a USB 3.0 drive or see if I can copy what's on my 2.0 drive directly onto the SD card, but this is mainly for testing anyhow. I imagine I'll make room for another OS directly on my SSD if this test goes well.

>> No.1608238

>>1607997
ooh ooh c gets brighter, a, b get dimmer

>> No.1608242

>>1608238
d also gets dimmer, genius

>> No.1608246

>>1608242
these little circuit problems remind me more of 4th-grade directions-following class than 9th-grade physics

>> No.1608316

>>1608193
I'm a big fan of KiCAD. If it had a good way of repeating layout for repeated schematic pages and a way of classifying nets and setting DRCs for different classes, I'd make it my company's default. Going to have to stick with Altium for those two features, though.

>> No.1608317

Is the traditional 817 optocoupler fast enough to handle audio frequencies with significant distortion?

I need an opto on the output of the high voltage diff pair to drive the class AB amplifier then I need another opto to provide the feedback to the HV diff amp.

>> No.1608331
File: 3.55 MB, 2304x4096, IMG_20190510_195514.jpg [View same] [iqdb] [saucenao] [google]
1608331

Hello guys, can you help me out? I have this flashlight that keeps discharging when not turned on.
I took it apart and found the problem will be on this circuit, but I don't know exactly where. When I connect the battery, it instantly starts to discharge, and the bug turns burning hot. How can I diagnose it further/fix it?
Thank you.

>> No.1608332
File: 1.78 MB, 4096x2304, IMG_20190510_195527.jpg [View same] [iqdb] [saucenao] [google]
1608332

>>1608331
Bottom

>> No.1608344

>>1608316
>If it had a good way of repeating layout
it has a hacky but effective way which helped my sanity a LOT when designing a multi-channel LED driver board
https://raw.githubusercontent.com/mmccoo/kicad_mmccoo/master/replicatelayout/replicatelayout.py
>and a way of classifying nets and setting DRCs for different classes
this. I really wish I could tell it that net class HV can be .016" from net class HV-power and net class signal can be .008" from net class power, but HV and HV-power have to be .060" from signal or power. I also wish design rules could be exported and imported into a board in case someone wants to change vendors on the double

>>1608317
maybe, but it's another phase shift which might affect your loop stability and should be accounted for

>> No.1608349

>>1608316
Am i going to be missing those features in simple Hobby projects?

>> No.1608352

>>1608331
So it's that 8-pin chip that gets hot, not the two power resistors on the back? She sounds dickered mate.

>>1608349
Repeating layouts can be pretty useful, but I've found somewhat clumsy ways of doing that that are pretty functional on KiCAD (i.e. editing one subsheet edits all the duplicates, but each one has different part identifiers so the BOM works out fine). Net command is also pretty clumsy in KiCAD, but I've never really needed to use it much. You should be fine.

>> No.1608355

>>1608352
Yeah, it's that chip.

>> No.1608358

>>1608355
Want to try looking for that chip's datasheet? I can't quite read the part number from that image. It might be possible to see where it's gone and broken, or at the very least figure out what sort of rudimentary circuit could be used to replace it.

>> No.1608359

>>1608242
Dim Bulb Award. D is not among "the other bulbs".

>> No.1608360

>>1608359
no, "the other bulbs" is referring to A, B, C, and D, with the actual bulb in question being the subject of the question, because the author is assuming you're an onion.

>> No.1608372

>>1608360
confabulation too lame, award sticks.

>> No.1608384

>>1608358
CS4956 SemiCoa

>> No.1608385

I am working on modifying some speakers, and I am trying to add a wire to a terminal post. The end where all the crossover wires go to has a strip of metal on it that has two holes, but for some reason one of the areas of the strip refuse to take solder. I tried adding some flux, but the solder still beads up and flakes off. Any advice for what I can do?

>> No.1608386

>>1608358
I tried, I wasn't able to find the chip.
How probable is it that the problem is in it? Couldn't something else pour more juice in it, causing it to warm up?

>> No.1608388
File: 87 KB, 759x761, 8899044961_1333967240.jpg [View same] [iqdb] [saucenao] [google]
1608388

>>1608386
The only datasheet i found about this Chink IC

>> No.1608391

When milling pcbs, how much lung cancer can you actually get when using no suction or mask?
Since you mill only like half a milimeter i can't see it being too much

>> No.1608392

>>1608386
>Couldn't something else pour more juice in it, causing it to warm up
It could be being overvolted, but I don't think that's very likely considering its power supply. Arguably one of the passives broke and it's got a MOSFET inside that is sinking way too much current (it has heat-sinking vias for a reason), so by doing some part replacement or removal it may be possible to run again. After all it hasn't gone open circuit yet.
If you're up to it, try tracing out the circuit and drawing it in an easy to view fashion, including whatever plugs into those three sockets.

>> No.1608396

>>1608388
Hmm, how/where did you found that?

>> No.1608397

>>1608392
I will try, but it will look like shit

>> No.1608403
File: 2.40 MB, 2304x4096, IMG_20190511_125813.jpg [View same] [iqdb] [saucenao] [google]
1608403

>>1608397
As i'am looking at it, I can say that I will make Out of it.
But I realized that I took a terrible photo of circuit. Left plug is battery, mid and right one are for different LED strips. 3 red cables are going into 3position switch(off, led1on, led2on)

>> No.1608406

>>1608385
>one of the areas of the strip refuse to take solder.

oxidation needs to be removed. start gently with an eraser and get rougher if that doesnt work: steel wool, sandpaper, a file.
could also be that your iron is under-powered for the job. you probably need 35W min to handle a typical terminal post.

>> No.1608412

>>1608396
>how/where
google/internet
looks a lot like a clone of the TP4056
https://dlnmh9ip6v2uc.cloudfront.net/datasheets/Prototyping/TP4056.pdf

>> No.1608416
File: 41 KB, 800x600, Spittlebug.jpg [View same] [iqdb] [saucenao] [google]
1608416

I just wanted to say that I found out that "Spittlebugs" can eat through hot glue from a hot glue gun and cause shorts in wires. I had a temporary line above ground that had a split that was tinned, soldered together, and encapsulated with hot glue. A short was detected in the line and it was eventually discovered that a spittlebug caused it. It ate little holes into the hot glue all the way down to the wires. I guess it was looking for a plant stem and plant juices. I had to re-splice everything.

>> No.1608422

>>1608416
Just put spiders on your wires so they eat the bugs, easy.

>> No.1608429

>>1608412
Tried to google it before, no results.
Thank you very much.

>> No.1608436

>>1608412
That link you posted is for the 4056, is there a link for the 4956? I had a good hard look for it also and came up with nothing. There was a chinese site that were advertising that chip, but without a datasheet.
Anyhow, a cursory look of the circuit is basically identical to how a TP-4056 is as far as layout/pinout goes, so if the package can somehow be desoldered it can be fixed with a drop-in replacement from alibay. But it's going to need a hot air gun.

>> No.1608449

>>1606586
I'm looking into making arduino midi controllers, I've got my eyes set on traktorino.
Any projects thst include jogwheels? I haven't found any and I'd be great to be able to have a jogwheel on my diy project

>> No.1608458

>>1608436
it was this page
http://wimtel06.sm160.com/Company/OfferDetail/1-000-5612-069.html

>> No.1608459
File: 23 KB, 330x357, pc817.png [View same] [iqdb] [saucenao] [google]
1608459

>>1608317
>fast enough to handle audio frequencies
yes, -2dB at 20kHz
>with significant distortion?
distortion is for free

>> No.1608460

>ordered €200 worth of parts
>DHL says they'll deliver today
>patiently waiting at home
>get email that they missed me
>didn't even ring the doorbell
Goddammit, I was looking forward to doing a project this weekend.
Lying delivery guys need to get the rope.

>> No.1608485

>>1608032
>If this is completely retarded please correct me
It's accurate enough that engineering programs that start general use pretty much exactly the same analogy.

>> No.1608499
File: 224 KB, 961x635, water_amp.png [View same] [iqdb] [saucenao] [google]
1608499

>>1608485
>>1608032

>> No.1608537
File: 3 KB, 419x125, ccaaa.png [View same] [iqdb] [saucenao] [google]
1608537

can someone tell me why the red section isnt floating?

>> No.1608545

>>1608537
because it is part of the black game

>> No.1608554
File: 485 KB, 1536x2048, IMG_20190511_190302.jpg [View same] [iqdb] [saucenao] [google]
1608554

My pcb mill attachment is in place
And now we wait for the pcbs to arrive
I wonder ho well it will mill with the diamond ball tip since it's the only one i have.
If it's shit, i'll have to shill out for something better

>> No.1608604
File: 51 KB, 958x340, Clipboard01.jpg [View same] [iqdb] [saucenao] [google]
1608604

Did you know that bendy plastic sticks are too dangerous for chinks to put on an airplane?
What the fuck did they made them out of? plutonium?
There goes my dream of having a flexible pcb holder soldering base ;_;

>> No.1608624

Can you recommend me some affordable and good enough capacitor meter ?

>> No.1608649

>>1608554
3d printers arent designed for lateral forces at all.
Doubt its going to work well, if at all.

You should just print a pen holder and make the 3d printer a plotter. Then just put those PCB etching markers on it, where you mark the traces in special marker then dissolve all the rest of the copper in acid.

>> No.1608651

>>1607291
The problem with the glowing light is I need it in my house. The fence is not even in sight, so I need some way of activating a relay with a 10000 volt signal. Might be an issue. I'm gonna use multiple DC bull fences. Most are rated for 30 miles but I want them to hurt and I need to know what side of the property they have been cut on.
I did find 900 mhz camera equipment, I think that will work, even with all the trees. I will look into a Omni directional antenna and mount it on a tower.

>> No.1608655

>>1608649
Already did that
>>1607921

That is a backup plan, since i want to avoid icky acid if i can.

And it's not like i'm going to mill 5cm deep into titanium, it will be 1mm end mill going .5mm into soft copper and i know from experience that the steppers on the printer are strong as fuck and will likely have no issues with this

>> No.1608661

>>1608655
>it will be 1mm end mill going .5mm into soft copper

nah. copper is typically only 1.4 mils thick. (a mil is one thousandth of an inch.) any deeper and you're digging into fiberglass or phenolic resin, which could cause a Mesothelioma type cancer if breathed in often.

makes the icky acid seem not so bad in comparison

>> No.1608667

>>1608624
The chinks sell them over at aliexpress and ebay for 5 to 15 bucks, depending on the model. I got one of these in the mail yesterday
https://www.aliexpress.com/item/GM328-Multi-use-Transistor-Tester-DIY-Kit-Diode-Capacitance-Voltage-Meter-PWM-Square-Wave-Signal-Generator/32829897587.html
There's a MASSIVE thread about them at eevblog. Turns out they're pretty good. Not as good as an expensive LCR meter, but good enough for hobby stuff and super useful.

>> No.1608668

>>1608661
pcbs aren't perfectly flat so i have to mill deeper to make sure i don't leave shorted traces behind
i will attach some dust ring thing around the end mill to block the dust and maybe buy one of those masks so it should be fine, it's not like i want to live to 90 anyway

>> No.1608730
File: 99 KB, 691x960, 8jtslhdjnmr21.jpg [View same] [iqdb] [saucenao] [google]
1608730

is there an easy way to convert a 8-bit parallel value to some pulse width modulation output? I want to read a value off an eeprom and control a servo with it. not with an arduino though, I mean more like with some IC that has it built in the hardware.

thanks

>> No.1608736

>>1608730
the easy way, of course, is with a microcontroller
it'll take a few chips just to get the EEPROM to start divulging its contents, even more so if serial. that said, once you have your data values all nicely framed in octets, you can use an 8-bit counter of your choice and a magnitude comparator ('HC688 or 2x'HC85) to do (edge-aligned) PWM

>> No.1608750

>>1608736
thanks, I'll check these

>> No.1608843
File: 7 KB, 400x400, tegaki.png [View same] [iqdb] [saucenao] [google]
1608843

>>1608730
An arduino basically is a microcontroller, albeit a pretty inefficient one. Not that it isn't a good way of going about this project, but if you're apprehensive about using an arduino, I doubt you'd prefer using a barebones avr.
I'd use an R2R resistor ladder DAC which feeds a normal triangle wave comparator PWM circuit. No buffering needed either, since the output will be going into a high-impedance comparator anyhow. So that's like 2 ICs, assuming you make the triangle wave oscillator with the second comparator and your resistors are in a single-package network.

>> No.1608874

>>1608460
FUCKIN hate that bro; Once i ordered some samples from Analog Dev, and after not arriving, it said it had been returned to sender. Fast forwards a month, apparently these fucks sent it to another state to check for illicit materials, and then re-shipped.. so stupid

perhaps try thinking up a cool fix for something in your house

>> No.1608918

>>1608874
>tfw going through this right now
>>1608604
I wish there were support groups for dealing with this trauma

>> No.1609071

>>1608459
I'm a retard that can't into words apparently. I meant without significant distortion.

>> No.1609137
File: 60 KB, 720x902, 1549201576277.jpg [View same] [iqdb] [saucenao] [google]
1609137

Why people always suggest to use analog dead time circuit instead of put it on MCU code????

>> No.1609155

>>1609137
>people always suggest
That's a statement. Any evidence?
When you pull the plug, the music stops. When you insert a delay (a phase shift, e.g. a filter) it continues, modified.

>> No.1609172
File: 65 KB, 500x618, 1515873342773.jpg [View same] [iqdb] [saucenao] [google]
1609172

how would one go about stealing from overhead powerlines to power a crypto farm
asking in the spirit of preventing crime

>> No.1609202
File: 25 KB, 400x454, richard_box_field.jpg [View same] [iqdb] [saucenao] [google]
1609202

>>1609172
Use the field.

>> No.1609277

>>1609202
>gets inverse cube attenuation
no thanks

>> No.1609390
File: 1.24 MB, 3264x2448, 1mF.jpg [View same] [iqdb] [saucenao] [google]
1609390

PCB prototyping with a laser cutter? Went pretty well if you ask me, though I forgot the IEC socket to test it with.

>> No.1609404

>>1609390
Well the IEC socket's holes were too far apart, but that will be an easy fix in the files.

>> No.1609405

>>1609404
Oh and it looks like my PCB rivets arrived, I'll test them out tomorrow.

>> No.1609421

>>1607899
>No it doesn't, there's no such chip on it at all. It just has the 5V linear regulator and the ATtiny itself, and a bunch of passives.
are you sure you aren't mistaking attiny with digistump attiny?

>> No.1609422

>>1609390
Why not use perfboard?
And don't remind me of lasers.
I wasted 60 bucks on a chink lase which worked the first time i tried it for 20 seconds and then later wouldn't turn on anymore for some reason
And of course no such thing as warranty with chinks

>> No.1609424

>>1609421
They have the same chip on them.

>>1609422
Holes aren't big enough for that cap, also the switch and IEC socket obviously have non-circular holes.
As far as the laser goes, are you sure you drove it properly with sufficient cooling? Last I saw the diodes need feedback of some sort.

>> No.1609427

>>1609424
>you drove it properly
I came with its own driver board

>> No.1609428

>>1609427
What about cooling?

>> No.1609429

>>1609428
It had a big metal case with aliminium ribs and a fan, it was a complete package you can start using.
I am not sure if the laser died, or the driver did.
I am not sure how to test the laser module without the driver board without destroying it, so i just tossed into a box to and tried to forget my horrible waste of money

>> No.1609432

>>1609429
Lasers are just diodes of a sort, so a simple very low current driver made with an LM317 should give you a measurable forward voltage. On the other hand, powering up the driver without anything connected to it is probably fine.

>> No.1609441

Just to be clear: if I have to measure Vgs on a jfet, I measure from gate to source, right?

>> No.1609458
File: 3 KB, 257x315, osc.png [View same] [iqdb] [saucenao] [google]
1609458

>>1609441
Yes, sometimes indirectly.

>> No.1609473

Anyone used the RCWL-0516 motion sensors?
How problematic they are with flase triggers?
They seem ideal for my project but it seems that wireless signals around them fuck their asses good

>> No.1609477
File: 392 KB, 2048x1536, IMG_20190513_135059.jpg [View same] [iqdb] [saucenao] [google]
1609477

what the fuck is this?!
voltage regulators FOR ANTS??????

it looked way bigger in the damn photos
this is fucking useless.
not even fucking amoebas would be able to solder this shit

>> No.1609478

>>1609477
fuck i meant mosfets not regulators

>> No.1609483
File: 291 KB, 640x480, 1557012687620.gif [View same] [iqdb] [saucenao] [google]
1609483

>>1609477
bruh

>> No.1609496

>>1609477
Isn't that just SOT23?

>> No.1609507
File: 392 KB, 1938x1595, question.png [View same] [iqdb] [saucenao] [google]
1609507

Don't know if /ohm/ is the proper place for this but I'm wondering if anyone could help me out with this question (EE student revising for comms exam).

I understand why sampling at 60kHz works (nyquist rate), but why does 25kHz work?

>> No.1609546

>>1608221
unetbootin used to be used when live images were CDs/DVDs that didn't include USB or other drivers active. unetbootin has been out of style for like a decade, because it actually fucked with the image (to install/redirect to the USB)

>> No.1609624

Can you recommend a decent microscope for under fity buckaroos? my eyes cannot deal with this smd shit

>> No.1609646

>>1609624
I use strong reading glasses (+4..7) because stereo, good field of view and headmount; also see >>1608808

>> No.1609653

>>1609646
>glasses
How does that work? if i put on someone's glasses everything is blurry as fuck

>> No.1609667

>>1609646
Watchmakers also have good magnifying glasses.

>> No.1609718

Can you solve this mystery.
>electric scooter
>drive normally (16-20 kph)
15 km range
ECU and wheel motor are cool
>drive like crackhead (25+ kph, throttle down)
23 km range
ECU is fucking hot, and wheel is hot too.

What the actual fuck?

>> No.1609727

>>1609718
Can wrong winding (let's say 1-2 wires going to wrong holes) cause this?

>> No.1609774

>>1609718
Uh, sounds like the ECU isn't very efficient at low speeds. But that extra energy has to be going somewhere, and if it isn't heat (because running it fast produces more heat than running it slow) then I haven't the foggiest.

>> No.1609836

>>1609774
>Uh, sounds like the ECU isn't very efficient at low speeds.
I thought the same but, it doesn't make sense, since ECU is cool, while in full retard mode shit is getting hot.
Btw, current controller is BLDC, but stock one treated motor as PMSM.
Also this controller runs great in sensorless mode with other motors, except this...

>> No.1609843

>>1609836
Kek. I though they are the same, but no...
I will rewind this shit.

>> No.1609850

>>1609836
All I can think of is that your method of measuring efficiency is flawed somehow. Either an issue with the battery voltage being reported incorrectly, your measurement of speed or time is wrong, or something else along those lines. Or possibly even a greater amount of heat is being produced in low-speed mode, but it's being produced where you don't notice it as much. Perhaps the bearing grease is too sticky at low speeds and causes them to be less efficient? It's shooting in the dark, but it is a very odd problem.

Also anyone know how you'd calculate the steady-state current of a PTC thermistor being used as a resettable fuse? I'm assuming it will oscillate a bit before sitting at a dozen µA or so.
And after a bit of googling around, I think putting the MOV before the EMI filter is the way to go. It protects the X capacitors from overvoltage, even if the chokes can handle it.

>> No.1609852

>>1609850
And this is a neat article on EMI filters, if you ignore the missing assets:
https://interferencetechnology.com/power-quality-effects-reliability-susceptibility-emi-filters/

>> No.1609886

>>1609850
>All I can think of is that your method of measuring efficiency is flawed somehow.
It probably is, because it makes no fucking sense. Even considering that PMSM is not ideal for BLDC controller. If it was that bad, it would heat like mad, but it doesn't.
> Either an issue with the battery voltage being reported incorrectly
Probably battery is being Chinese piece of junk, that has insane self-discharge. I also don't have any instrumentation, so I have no idea about voltage under load.

Anyway, I think I want more speed. Can I switch Y connection to Δ in BLDC motors?
I should have 1.7 times more speed, right?

>> No.1609891

>>1609886
A cursory search shows that it probably is possible, but I think the rewiring of the winding themselves could prove difficult.
Also please post an image of the internals of your scooter wheel, I'd love to see what sort of internal gearing they have, if at all.

>> No.1609899
File: 98 KB, 600x800, windings.jpg [View same] [iqdb] [saucenao] [google]
1609899

>>1609891
>Also please post an image of the internals of your scooter wheel, I'd love to see what sort of internal gearing they have, if at all.
No gearing, direct drive. Winding looks the same, as BLDC winding.
Photo is from the internet, welds on axle are not original. My stator is less skookum, it is thinner (30-ish mm vs 25 mm) and has different number of turns (lower speed, lower KV), and lower power... I genuinely thought they didn't bother making two different motors for high-end and low-end models.
Btw, previous scooter I had, it had a hand-wound wheel, it was ugly, but it worked really well for some reason, meanwhile this loooks tidy, but in fact pretty shit.
> but I think the rewiring of the winding themselves could prove difficult.
I know... This is why I'd try delta connection first.

>> No.1609905
File: 107 KB, 960x529, large.jpg [View same] [iqdb] [saucenao] [google]
1609905

>>1609899
>No gearing, direct drive
Ooh, that's interesting. The bike hub motors have a planetary set in them (pic related), but I guess since the wheel diameter is so much smaller you don't need it. I'd rather have an image of the rotor since that's where the gearing would be, but if you've looked inside your motor I suppose your word is good enough thanks. I can't really find any good images of the small scooter motor internals when I search.

>> No.1609906
File: 367 KB, 581x388, wheelie.png [View same] [iqdb] [saucenao] [google]
1609906

>>1609899
https://www.youtube.com/watch?v=eG0MJQOvydw
Dude disassembles wheel.
https://docs.google.com/document/d/1U1O4cMiKAlb62qDMkQ_rrFy15Pw4yJySlMJgjUY3Roo/edit#
More shit about shit.
Again, Ninebot has aluminium wheel, and magnets glued to it. My previous scooter was more skookum, it had steel ring and much thicker aluminum covers...
It feels like penis in a mouth, actually. Noname chinkshit is better than much more expensive brandshit. Chinkshit has normal BLDC controller (like I bought), no bluetooth botnet, regenerative braking turns on only when you brake (not like on Ninebot shit, where you release throttle and shit starts killing battery with regen for no good reason), more skookum parts...
>>1609905
>The bike hub motors have a planetary set in them (pic related)
Not all, there are a lot of bike motors with direct drive. Chinese magnets are quite strong.

>> No.1609908
File: 174 KB, 800x600, motor-kolesa 01.jpg [View same] [iqdb] [saucenao] [google]
1609908

>>1609906
Here is ebike DD motor. Same structure, but more thickness, more windings, more torques.

>> No.1609959

Just measured the propagation delay of my shitty BJT inverter at 875ns with my new logic analyser. It's also the only thing that's working properly in my circuit, as I tried to build a 4-stage frequency divider with 74HC164 shift registers and I'm getting a division of 8 with the first two stages (instead of 16) and no division in the next stages. I think I'm getting glitches out the wazoo, I guess a breadboard isn't a very good medium for even 50kHz without any capacitors.

>> No.1609986

>>1609959
Sure you're not doing it very wrong? It's pretty easy to get 1MHz+ oscillations from ring oscillators.

>> No.1609988

>>1609986
I don't think the shift registers are the problem, but rather the lack of ripple-suppressing capacitors. My RC oscillator is also suboptimal, I should probably have used higher impedance passives.

>> No.1610019
File: 24 KB, 353x353, 1536034237584.jpg [View same] [iqdb] [saucenao] [google]
1610019

>>1609850
>resettable fuse
how much voltage are you putting across it that ~10uA produces adequate I^2R self-heating to keep it in the tripped state?
fuses are usually designed-in to protect wiring from malfunctioning equipment. you might want to look at some polyfuse data sheets to see what you can reasonably expect from them
thanks for good article on EMI filters fren, have a cat

>> No.1610029

>>1610019
Mains voltage. I guess it's more likely to be 10mA though.

>> No.1610058
File: 354 KB, 1536x2048, IMG_20190514_135637.jpg [View same] [iqdb] [saucenao] [google]
1610058

Care package from Ping Pong is finally here. But now I am faced with a great dilemma.
Do I mill, or do I dissolve?

>> No.1610063
File: 320 KB, 1536x920, 1530762594229.png [View same] [iqdb] [saucenao] [google]
1610063

>>1609988
shoot-through is when H-bridges (or totem-pole outputs or CMOS logic cells) have both their pass elements turned on at the same time as they transition from one state to the other, therefore behaving briefly as a short to ground. think of those decoupling caps on logic devices as local, low-inductance sources for shoot-through energy and the gate charges being moved to and fro. and think of those wires (and component leads) as inductors (and antennas). the natural consequence of such thinking is to put those supply caps right on top (or under) the chip in question with the shortest leads practical, Pic related

>> No.1610105

Seeing as how this thread is close to finishing(?) I just have a question for the regulars here, how long did it take for you guys to get this 'experienced?' This is from a guy who starts his EE degree next year and has no idea what's going on but want to join. Did everyone just follow the book recommendations in the Principle section? Sorry if I sound confusing

>> No.1610107

>>1610105
I've been fucking around with this shit for like 8 years now. Learn by doing.

>> No.1610111

>>1610105
I've been fucking around with this shit for like 28 years now. learn by doing

>> No.1610113

Regarding distortion circuits: when doing asymmetrical hard clipping with diodes, does the specific ratio of forward voltages have an effect when it comes to harmonics and stuff?

>> No.1610138

>>1609718
Sounds like it's running pwm, and at higher speeds, your duty cycle is mostly "on" so it's gonna be hot

>> No.1610140

>>1610105
Just like the others said, been doing this forever. I loved to take shit apart as a kid and see how it worked. Trial and error, don't be scared to break something, but always learn to fix it.
Did 4 years and got an ee degree, really helped out, but I've still been doing my own research since graduating.

>> No.1610199

>>1610140
Did having prior experience in doing this massively help you in EE? Like is it closely tied to your EE materials? Cuz I'm essentially following your footsteps as I'll be starting projects over the summer to get some firsthand experience

>> No.1610207

>>1609908
>>1609906
>>1609905
>>1609899
>>1609886
>>1609836
Okay, mystery solved.
Motor windings have relatively high resistance (unlike other motors I had). In full retard mode in 99.9% current is below limit, so controller basically works as commutator.
In low-speed mode it actually uses PWM, but due to lower current nothing heats that much.

>> No.1610244

>>1610105
I have an Electrical Engineering degree and I'm not very good, though most of it is depression and had to quit a good job.

>> No.1610253

>>1610207
Does that make sense? There's nothing inherently inefficient about using PWM in a BLDC, is there? And the energy still needs to go somewhere.

>> No.1610361

>>1610253
>here's nothing inherently inefficient about using PWM in a BLDC, is there? And the energy still needs to go somewhere.
You have switching losses (they are much higher, than plain 100% ON), and it is heat.
Since current is lower, it is not noticeable (controller has decent heatsink, couple watts is nothing for it), but overall efficiency is lower.
When you go full retard, more power is lost, but overall efficiency is higher.

>> No.1610393
File: 8 KB, 400x400, tegaki.png [View same] [iqdb] [saucenao] [google]
1610393

>>1610361
>drive normally (16-20 kph)
15 km range
ECU and wheel motor are cool
>drive like crackhead (25+ kph, throttle down)
23 km range
I guess that's feasible. I think I misinterpreted it as "drive normally = lasts for X minutes", "drive fast = lasts for Y minutes" where X < Y, which I think would be impossible with the switching losses explanation. He was talking about the controller temperature, which would be roughly proportional to power waste, which is energy wasted per unit time, it just made sense for time to be the subject as opposed to distance.

Makes you wonder why they PWM at such a high frequency. Couldn't they instead just use a lower frequency or decrease the on-time? Though I'm hardly an expert on how these motors are driven, and smoothness of motion would need to be ensured. I'm assuming they don't try to match a sine to them?
The Lime scooters have an on/off throttle instead of a gradual one, so perhaps it's for the best.

>> No.1610401

>>1610393
First, I might have different load (i.e. wind). So I need further tests. But as far as I remember, weather was more less the same.
>Makes you wonder why they PWM at such a high frequency.
In order to have more control over current, since load is inductive. I don't know inductance, but I'm sure that 8 kHz is minimal frequency, otherwise you would blow the fuck out of fets.
>Though I'm hardly an expert on how these motors are driven, and smoothness of motion would need to be ensured. I'm assuming they don't try to match a sine to them?
Mass is smoothing them. BLDC (trapezoidal dumb controller I have) will always have torque ripple.
Sine controllers exist. Some of them are true FOC control, they provide more torque, more power, they are supposed to be more efficient, and quieter. Those fuckers need to know inductance of motor in order to use it. Some sine controllers are similar to BLDC in terms they will drive anything, and they have no benefits besides quieter operation.
I'm not an expert too...
Also, this controller uses resistors, transistors and diodes in order to drive FETs, not designated IC.

I think, motor is just shit, has wrong inductance or something...
>The Lime scooters have an on/off throttle instead of a gradual one, so perhaps it's for the best.
Really? Fucking deathrap. You need to have at least 1 liter of beer in order to drive in this crackhead mode, otherwise you will crash instantly

>> No.1610409

>>1609507
25k fs works in this case because the alias is so low frequency it doesn't overlap the original spectrum. So you can recover it with a bandpass filter, just like the answer says

>> No.1610437

>>1610244
I'm coming around corner on degree, and I hate to say you don't learn shit in college but its kinda true. I failed electronics I because I the whole time I was thinking "what the FUCK are we amplifying with this transistor we keep perfectly biasing, what if we don't, what's on the input, why ???" and don't even get me started on hybrid pi. When we started actually amplifying signals (v/i, duh) I was too far gone. ultimately my hobby interests shifted from embedded to analog a year ago and haven't looked back. I recommend it to keep things fruitful. because ultimately that's all that was missing, application. For simple circuits, it all seems way over the fucking top, but after you learn the simple circuits, the theory becomes very helpful if not essential. So my advice: start working on simple circuits, interfacing with a micro helps and makes things more fun til you can do what you need without it. Keep real simple, when you get bored or annoyed you can't do more, then you'll develop the fire for your more complex classes in a year or two

>> No.1610438

>>1610437
fuck, meant to end with. Learn by doing.

>> No.1610444

>tfw i watched one tutorial on kicad and was able to make a pcb all by myself after that
feels good to be a cad prodigy

>> No.1610446

>>1610444
Same, it's certainly one of the better FOSS alternatives out there. Inkscape and GIMP are so clumsy compared to Illustrator and Photoshop.

>> No.1610447

>>1610446
i use windows and photoshop at home and gimp with linux at work and gimp is a fucking nightmare. it takes me FOREVER to find how to do something that is completely intuitive and simple in photoshop

>> No.1610449

>>1610447
GIMP simply doesn't have a lot of the features PS has. But Inkscape could have most of the features of Illustrator, but I just can't find them.

>> No.1610478

If a 12ax7 electronic tube has a resistance of 60kohm and a maximum voltage of 300V (according to google), does it mean that it can't amplify a signal to more than P = U*U/R = 1.5W ?
Is that a lot or very little for a radio transmitter ?

>> No.1610480

can I make a PCB with MS paint

>> No.1610482

>>1610480
sure
http://www.zen22142.zen.co.uk/Prac/vero/vero.html

>> No.1610484

>>1610437
>interfacing with a micro helps
this. in most cases you're more interested in having a stimulus than a pristine, organic, small-batch handcrafted signal, and when you do need the latter, still nothing stops you from making one

>>1610478
>can't
correct
>a lot or very little
1.5W peak power is fairly small. it's about family walkie-talkie power

>>1610480
sure but it'll take you over 9000 hours

>>1610446
>one of the better FOSS alternatives
I think this is 80% because CERN took the project under its wing and people who have worked with commercial software in the eCAD space have a better perspective of what can be done and what makes sense in the context of the design workflow, vs. what's a "cute" research feature

>> No.1610487

>>1610478
>>1610484
I see, thank you for confirming my doubts.

>> No.1610490

Any chemists here?

If i mix baking soda into fecl3, is it safe to pour it down the drain after that without it melting my pipes?

>> No.1610494

>>1610490
>without melting pipes
probably
>safe
no, you'll be introducing significant copper into the waste water system which is a no-no. better to evaporate the neutralized solution and place the solids into a plastic bag to go into your normal trash collection
also next time use CuCl etchant so you aren't in the side business of toxic waste disposal

>> No.1610496

>>1610494
>significant copper into the waste water
it's not like i am the one who is going to be drinking that later

>> No.1610523

>>1610494
>needing H2O2
>not using aqueous iodine or bromine that turns into a solution of copper halide that you can simply electrolyse back again

>>1610496
It's a biohazard because it isn't treated properly by the waste water treatment plants. Bacteria can't "break down" copper ions, that shit goes straight to the ocean or rivers and isn't exactly friendly to the wildlife.

>> No.1610531

>>1610494
>>1610523
Is it possible to use the used ferric chloride for electroplating stuff with copper?

>> No.1610533

>>1610393
> Makes you wonder why they PWM at such a high frequency. Couldn't they instead just use a lower frequency or decrease the on-time?
Motors are current-driven devices. Typically, the controller switches the voltage to maintain a desired current waveform in the winding. The pulse width depends upon the winding inductance; if the inductance is too high, it will limit the achievable current at high motor speed (when the desired current is at its highest but so is the back-EMF).

So the inductance needs to be low enough that the current can ramp up and back down within the commutation interval even when the desired mean current is high and the commutation interval short. But that means the pulses have to be short to avoid exceeding the peak current. At low speeds, the desired current is lower, the back-EMF lower, and the commutation interval longer. And you can't just have one short pulse per commutation interval because that's going to vibrate like crazy.

Also, efficiency favours continuous-conduction (inductor current remaining roughly constant, i.e. fast switching) over discontinuous conduction (letting the current fall to zero between pulses). Mean current is the integral of current, copper loss (I^2R loss) varies with the integral of current squared. Halving the duty cycle but doubling the peak current gives the same mean current but doubles the copper loss. If most of the loss is in the motor (which is likely), reducing copper loss at the expense of increased switching loss in the controller is a net win for total power efficiency.

>> No.1610539

>>1610523
>isn't exactly friendly to the wildlife.
This is the future generation's problem

>> No.1610541

>>1610539
this is what happens when you don't exterminate sociopaths on sight

>> No.1610545

>>1610531
Something comes out as a precipitate (metallic iron?) and that's never good for trying to have a reversible system. But getting a fully used-up solution (basically copper chloride) with all the precipitate removed and you probably could use it to electroplate, so long as you avoid the production of chlorine (or any other gas).

>>1610539
And your problem too if someone finds out

>> No.1610548

>>1610545
>And your problem too if someone finds out
and who would that be exactly, when everyone's dead from copper poisoning?

>> No.1610549

>>1610548
>everyone's dead from copper poisoning
I doubt any human would die, but the local council would be pretty mad that the waterways were undergoing some very unnatural changes. Arguably the copper could bioaccumulate in eels or salmon or whatever and get eaten by humans and give them some copper poisoning, tipping off the authorities in the first place, but probably nothing life-threatening.
Anyhow, copper is valuable, anyone just dumping it down their drain is a fool, they should replace it with a cheaper but more reactive metal in a simple substitution reaction if electrolysis is too difficult.

>> No.1610571

>>1610545
I found a patent about regenerating etchant using electrolysis: http://www.freepatentsonline.com/4396475.pdf

If I understand it right, the cathode gets plated with copper and the Cl ion binds with Fe(II)Cl again to turn back into Fe(III) chloride.

>> No.1610583
File: 55 KB, 500x375, 6284a3051b273d78d2a04c6c12383c22.jpg [View same] [iqdb] [saucenao] [google]
1610583

I'm about to have work done on my workspace's roof so i't s the perfect opportunity to remodel/restructure.
I'm currently looking at an empty 12m x 5m room, what should I do with it? Any cool workshop pictures I could use as inspiration?

>> No.1610594

NEW THREAD
>>1610590
>>1610590
>>1610590

>> No.1610676

>>1610437
My problem is partially I'm poor as fuck and unmotivated. I don't have the money to get a dozen arduinos, even if they're $5 a piece, and I don't want to sit there soldering 50 components so I can make a toy. If I had my own home it'd be a different story, as well as if I had a job in which things would be relevant. There's a few applications I have for analog devices, but besides the "not my home" part, the other part is they're... soft applications, i.e. to the human body. It takes conductive gel, sitting in place for a while, other annoying things to do by myself. It's not an exaggeration to say I've been hospitalized repeatedly for depression, so I have very little ability to do anything repeatedly and consistently.