/ohm/ - electronics general, solder is for suckers edition

[Anonymous No.1438894]

dried-out thread: >>1434867
>I'm new to electronics, where to get started?
There are several good books and YouTube channels that are commonly recommended for beginners and those wanting to learn more, many with advanced techniques. The best way to get involved in electronics is just to make stuff. Don't be afraid to get your hands dirty.

>Books?
Beginner:
Forrest Mims III, Getting Started in Electronics
Charles Platt, Make: Electronics
Michael Jay Geier, How to Diagnose & Fix Everything Electronic

Intermediate:
Kybett & Boysen, All New Electronics Self-Teaching Guide
Paul Scherz and Simon Monk, Practical Electronics for Inventors

Advanced:
Paul Horowitz and Winfield Hill, The Art of Electronics

>YouTube?
mjlorton
BigClive
paceworldwide
eevblog
EcProjects
greatscottlab
AfroTechMods
Photonvids
sdgelectronics
TheSignalPathBlog

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

>Components/equipment sources?
Mouser, Digi-Key, Arrow, Newark are global full-line distributors with small/no minimum order.
RS Components (Europe)
eBay/AliExpress sellers, especially good for component assortments/sample kits (caveat emptor)
Your local independent retail electronics distributors
ladyada.net/library/procure/hobbyist.html

>Circuit simulators?
This mostly comes down to personal preference. These are the most common ones though:
LTSpice
CircuitJS (quick, dirty, interactive, web-based)
NI Multisim
CircuitLab
iCircuit for Macs

>PCB layout software?
KiCAD (recommended), why use anything else

>My circuit doesn't work. Halp?
Check wiring, soldering, part pinouts, and board artwork if applicable, then post schematic. Supply ALL relevant info and component values when asking for help.
>Li+/LiPo batteries
Read this fine 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.

[Anonymous No.1438918]

blah, so after much sock-puppet lobbying the fat useless slog BigClive has made it into the sticky. sad.

[Anonymous No.1438921]

>>1438918
it was like that when I got here. if enough people bitch I'll yank it

[Anonymous No.1438961]

>>1438921

He's my favorite youtube expert. Move him to the top of the list.

[Anonymous No.1438999]

I have pic related. It uses a transformer, it's an AC welder. I want to convert it to DC. I know I need a bridge rectifier and a couple caps but I don't know how to determine what value my caps should be. Open circuit voltage is 27v but I don't know how much that will jump up after rectification(I would assume somewhere around 34v). I imagine something in the 50-60v range would be more than adequate. How would I determine how many mfds to I need to properly smooth out the wave?

As for the bridge rectifier I imagine anything rated for more than 50 volts/150 amps with a spot for a heat sink would work. I watched a video of one guy doing this and he used an inductor in row with the gun, I don't know anything about inductors but would this be a neccesary step/good idea? Everything he did seemed overbuilt but it did work like he wanted.

[Anonymous No.1439039]

>>1438999

My Harbor Freight welder failed on me yesterday for the very first time. Armed with intimate knowledge of the inner workings of my unit, I did not automatically junk it and go buy another like others. Instead, I opened it up and determined the problem.

The welder was welding particularly cold, and I saw smoke coming from the torch handle. Eventually it failed to arc at all when I pulled the trigger; it only fed the wire. So I touched the wire to the ground clamp and pulled the trigger. Only wire. Then I opened the case, released the wire feed roller, and touched the wire to the positive pole on the capacitor. Trigger pull; no arc. I took a multimeter, set it to Volts DC and tested the poles of the capacitor. It charged up to 30 some volts, so I knew the innards of the welder were working fine, and that there was a fault between the capacitor and the torch tip on the negative side. I opened up the torch and saw smoke residue inside, and where the lead was supposed to be electrically connected to the brass torch tip there was a lot of nasty looking stuff. I pulled at the lead and saw where it was supposed to be threaded, and immediately knew that thar was the problem. So I sanded the surfaces and soldered the connection together, assembled the torch back together, and proceeded to continue welding on my merry way. No cost repair. Powerful knowledge.

http://forum.driveonwood.com/t/modifying-the-harbor-freight-120v-welder-if-you-already-own-one/64/5

[Anonymous No.1439049]

>>1438999

to convert RMS to peak voltage, multiply by 1.41. so 27Vac gives 38Vdc. add 25% safety margin and you get 47V for the caps. available are 50V or 63V ones.

for reasonable smoothing, calculate 1000uF per amp, so for 125A, you need 125,000uF which will cost about about $80 at Mouser. i would try it without caps. from my limited knowledge, any differences will be subtle.

[Anonymous No.1439061]

>>1439039
I'll look through that. The Flux 125 replaced the older Flux 90(the machine being discussed there) but as far as I've been able to tell the design has been altered to feature newer parts. It seems to be in large the same "90amp" welder with claims of hitting 125 amps. I'm sure it's applicable. Thank you.

>>1439049
I always forget that it's even refered to as RMS, my grasp of theory is sub par at best. That's a ridicluous value for the cap, I imagine it's the size of my fist. Aside from price I think that I'd have a fun time trying to even fit it. Do you think parrallel caps might be a more sensible solution. I will try it without the caps if you don't think it'll be a big deal. This is just the welder I'm learning on, eventually I'm going to get a better one. It doesn't need to be perfect, I just like tinkering with stuff and I've always heard dc is better for everything aside from aluminium tig.

[Anonymous No.1439082]

>>1438999
>he used an inductor in row with the gun
the inductor is more important than the capacitors

[Anonymous No.1439136]

>>1439082
>the inductor is more important than the capacitors

not really. also, making an inductor that can handle 125A and has enough inductance is a nightmare. not to mention that, being in series with the load, it'll decrease output current and get hot hot hot.

[Anonymous No.1439153]

>>1439049
>>1439061
It's an inverter. Most likely it outputs square wave, as producing square output is easier and works better than sine output. Thus, no need to multiply anything.
Also, the point of DC welding is that the direction of the current does not change. You don't really need to smooth it, albeit adding some capacitance might make the arc a bit more difficult to extinguish.

[Anonymous No.1439155]

This is stuff for 225 A. Old transformer core for the coil, 4 Diodes on heatsinks, fans, no capacitor required.

[Anonymous No.1439168]

>>1439136
I probably should have said,"Just as important."

A microwave oven transformer core - even a small one, makes a great welding inductor.
Just wind a welding lead through it as many turns as will fit - it won't be many, and it will provide the necessary inductance with almost no loss in power.
>making the welding lead a couple of feet longer with the same gauge wire won't change the current delivered.

[Anonymous No.1439177]

>>1439153
>You don't really need to smooth it
If you don't smooth it, the arc actually stops and starts sixty times per second.
Without smoothing, it's almost as bad as straight AC.
I still have the same Lincoln AC-225 I've had for forty years.
I temporarily added diodes to the output without capacitors or inductor and it was almost the same as not having diodes.
It still welds great for what it is but for really nice welds I use my DC inverter welder.

[Anonymous No.1439184]

>>1439177
>stops and starts sixty times per second.
That would be 120 times, but more importantly, inverters aren't limited to mains frequency output and the "pause" is much, much shorter than in your old welder. So, even if you add a filter capacitor, it can be significantly smaller.

[Anonymous No.1439203]

>>1439184
>That would be 120 times
if it's full-wave rectification
half-wave is 60 times

[Anonymous No.1439230]

>>1439153
>its an inverter
How can you tell? It looked like a beefy transformer in the gut shots I've seen. I haven't yanked mine apart yet though. Most of the cheap welders that use inverters that I've seen wear it like a badge of honor, having it listed everywhere.

>>1439082
The guy I found bought a ferrite ring off ebay and just (loosely) wrapped some 10 gauge wire around it and put it between the capacitors he used and the lead to the gun. Would that actually work? He claims it does but it could be confirmation bias. Whenever I've seen inductors in radios it's always been tightly wrapped enamel wire. Admittedly I don't know as much as I'd like and I apologize if this is crossing into spoonfeeding.

[Anonymous No.1439248]

>>1439230
>How can you tell?
Well, I can't. You're right, it has an old-fashioned transformer.

[Anonymous No.1439268]

>>1439248
Do you happen to know why it is called 'inverter' rather than converter? Can you invert energy?

[Anonymous No.1439270]

>>1439230
>10 gauge wire
There is nothing wrong with using ferrite core.
Being loosely wound isn't a problem unless you're trying for maximum turns.
Using plastic or tape insulated wire is OK too if the desired number of turns is achieved with the thicker insulation.
Using #10 wire at high amps and low volts will cause heating of the wire and reduced voltage and current at the electrode while the arc is active.
Using his core and the largest gauge wire that will allow the same number of turns through the core will give the same/similar inductance with less loss.

[Anonymous No.1439271]

>>1439270
>high amps and low volts
the reason I said low volts is because at 30v every volt counts

[Anonymous No.1439278]

>>1439248
Ok, I appreciate you confirming that.

>>1439270
>>1439271
Is their anyway to determine how many wraps would be ideal? What gauge of wire would you recommend? I can feel through the sleeve leading to the gun it's not much thicker than that. I'd have to assume that's one reason why the duty cycle is so low on this machine(only supposed to use it at max for 2 out of 10 minutes). The ground clamp on this thing is the same you'd find in $8 jumper cables.

[Anonymous No.1439283]

I want to make a fun little device with mechanical keys that plays tones when you push them. What IC should I use to produce these sounds? Im basically looking to make a crude toy handheld synthesizer.

[Anonymous No.1439284]

>>1439278
>Is their anyway to determine how many wraps would be ideal?
You'll have to trade-off the number of turns and the size of the core with the amount of space you have available or are willing to make room for.
(more turns on bigger core is better)
>What gauge of wire would you recommend?
use the same gauge as the ground lead for minimum loss.

[Anonymous No.1439289]

>>1439284
Taken to heart, thank you. I'll need to crack it open this week and measure available space. I've tracked down a 150amp and 200amp rectifier for under $20, quite a few larger ferrite cores for under $15, if I decide to try it a 200k micro farad can cap for $40 with shipping, and I have tons of heatsinks, wire and connectors laying around already. Available space will determine which rectifier and ferrite core I get.

Thanks for all the guidance and spoonfeeding.

[Anonymous No.1439300]

>>1439289
>The guy I found bought a ferrite ring off ebay and just (loosely) wrapped some 10 gauge wire around it and put it between the capacitors he used and the lead to the gun.

http://forum.driveonwood.com/t/modifying-the-harbor-freight-120v-welder-if-you-already-own-one/64
>Unfortunately I clamped my coil into a vise while I was making it, and beat it with a hammer. I broke the toroid core in two, which greatly diminishes the effectiveness of the inductor. It is difficult to wrap 4 gauge solid wire tightly around the toroid core, and I recommend not hitting it with a hammer while clamped in a vice. I went ahead and finished making mine with the cracked core.
> #4 solid wire

[Anonymous No.1439302]

>>1438894
What's the best software for making wiring diagrams/schematics for 3phase and 120v control circuits?

[Anonymous No.1439305]

>>1439300
That's not the guy I saw use the ferrite core, it was some dude on youtube who used something much thinner than #4. I definetly never thought that hammer was a feasible option. I haven't had a chance to read through that thread yet.

[Anonymous No.1439306]

>>1439305
>That's not the guy I saw
I understand
I was just pointing you to someone who seems to want a better outcome.

[Anonymous No.1439308]

>>1439306
I'll look into it. I have a lot of wire ranging from 4/0 braided aluminum to 36 gauge single strand. I was thinking that I'd just go through and see the largest that I could actually wrap. I'm trying to fit the largest core I can.

[Anonymous No.1439317]

>>1439308
Sounds like a plan, anon.
Keep us posted.

[Anonymous No.1439322]

>>1439283
there are many such designs based on the 555 timer IC

>>1439302
meh, we mostly deal in small signal shit here. "electrical cad" is a good search term to find programs specifically adapted to industrial electrical applications, if no panelfag chimes in with a more experienced recommendation

>>1439289
be sure to check the datasheets on those cores and make sure they're designed for 60/120Hz power applications, not RF. also, too many turns will saturate the core and waste energy in heat, which could impose its own duty cycle limitation

[Anonymous No.1439359]

>>1437727
>You're looking for the LT3080. That can go down to zero volts. True zero. Not very close to zero. Rather than having a voltage reference it uses a precision current reference instead and there's a way to set it for the voltage you want with external resistors or some shit. Read that datasheet. If for whatever reason you can't make do with that you probably want some switching regulator, buck-boost. Honestly you're probably better off with this for a PSU than a linear regulator based designed but it is slightly more difficult to design and switching regulators are noisier but if noise is a big concern just whack a 12-pole LC filter on the output and she'll be right.

I was trying to remember that one - but didn't Dave Jones use it, and people in the comments and on his forum point out some problems with it?
If i'm using it as a BJT driver, then I suppose it only needs to go as low as ~0.7V anyway

[Anonymous No.1439448]

>>1439308

my guess is that 4 turns of wire around a big ferrite core will produce very roughly 10uH. if you're working at 100Mhz, that'll have a significant effect, but at 120Hz, it's like it's not even there.

[Anonymous No.1439476]

Turning a transformer into an electromagnet, the coil on the right is connected - will removing the left coil make the magnetic field stronger? I can't simply increase the voltage as I only have a 12V supply

[Anonymous No.1439478]

Turning a transformer into an electromagnet, the coil on the right is connected - will removing the left coil make the magnetic field stronger? I can't simply increase the voltage as I only have a 12V supply

[Anonymous No.1439485]

I find the explanations really lacking in Art of Electronics whenever they present a basic circuit then "improve" it.
What other books would you guys recommend for exposure to circuit designs?

[Anonymous No.1439487]

>>1439478
>will removing the left coil make the magnetic field stronger?
if it's not connected or shorted, it doesn't matter whether it's in the field or not

[Anonymous No.1439493]

>>1439487
Ok thanks. I'll try scrounge up a higher voltage source

[Anonymous No.1439494]

>>1439478

seems to me that a transformer is the wrong form for making a strong magnet since transformers are specifically designed to keep the field inside the metal parts. you would probably want to emulate a solenoid or speaker design instead. i.e. cut off the top of the transformer with a hacksaw.

[Anonymous No.1439495]

>>1439494
how about delaminating the core and only using one set of the E profile sections? Just not sure how I would re-laminate the whole thing

[Anonymous No.1439496]

>>1439495
so I end up with something more like this

[Anonymous No.1439497]

>>1439495
>>1439496
force out a couple of the I's
then work out a couple of E's
then remove all the I pieces and half the E pieces (all E's from one side)
re-insert the E's to the same side as those remaining

[Anonymous No.1439504]

>>1439497
What glue should be used for this? Would it be wise to use a slip of tissue paper or something between the laminations to ensure insulation? Though I guess if you're just using it with DC eddy currents don't matter in the slightest.

[Anonymous No.1439505]

>>1439497
Thanks, I tried but because it's welded together along the base I'll have to grind the weld off and disassemble the whole thing

[Anonymous No.1439508]

>>1439505

why not use the much simpler, much quicker method i suggested above, cutting it with a hacksaw? it's mild steel, it cuts like butter.

[Anonymous No.1439510]

>>1439508
Well It's not "mild steel" but it is a steel that is mild.

[Anonymous No.1439518]

How much current does it take to turn on a solid state relay?
To my disgust i found out the normal relay i am using is wasting 60ma to keep itself on

[Anonymous No.1439522]

>>1439518
on the order of 10mA
but your "contact" losses on the other side will be significant, which is why SSRs of much current handling capability require hefty heatsinks
depending on what you're switching, reed relays or latching relays may be a worthwhile option

[Anonymous No.1439525]

>>1439522
>latching relays
But they only have two legs for the trigger signal, where do you attach the load?

[Anonymous No.1439526]

>>1439525
*reed relays

[Anonymous No.1439527]

>>1439518

you gotta be one cheap jew to begrudge a relay such a tiny amount of power to do its job. fucking slave driver!

[Anonymous No.1439529]

>>1439526

you're thinking about reed switches, not reed relays, jew!

[Anonymous No.1439530]

>>1439527
I don't mind paying for the few extra ma, the problem is that my shit is battery powered, a small battery at that, and 60ma is a lot of drainage. Just keeping the relay on with rest of the circuit off would drain my source in a day

[Anonymous No.1439532]

>>1439530

if you have one of these old thermostats, they will usu. have 2-3 latching relays. not rated for much load current, tho. they're like $3 at the thrift store.

[Anonymous No.1439535]

>>1439532
Yeah the latch relay seems like the solution i was looking for

[Anonymous No.1439538]

>>1439525
some make you reverse current through the single coil to move the contacts to the other state. others have one coil for each state (but still polarized)

[Anonymous No.1439549]

>>1439538
How the fuck do i reverse current? I know i can just switch the wires, but how do i do it with just transistors and shit?
Also i really hate looking for the electrical components in the local store catalog, since they obviously list them translated to my shit language and it always takes me fucking forever to find the correct english translation for it.

[Anonymous No.1439551]

>>1439549
wait op amps can reverse current.. but their output is shit, they would probably fry when they triggered the realy

[Anonymous No.1439552]

>>1439549
>How the fuck do i reverse current?

H-bridge.

[Anonymous No.1439555]

>>1439552
>H-bridge
I see, it's like bridge rectifier for DC

[Anonymous No.1439557]

>>1439555
>I see, it's like bridge rectifier for DC

not in the least.

''An H bridge is an electronic circuit that switches the polarity of a voltage applied to a load. These circuits are often used in robotics and other applications to allow DC motors to run forwards or backwards.[1]'' -- wikipedia

[Anonymous No.1439558]

Why does my diff amp output always saturate to VCC-Vbe no matter how low a signal voltage I feed in?

[Anonymous No.1439563]

>>1439557
Well since i am going to use that relay as a breaker i just realized i don't even need to reverse the polarity, since when the relay switches i will want to reset it manually anyway

[Anonymous No.1439564]

>>1439549
I've seen designs that use logic gate outputs on each side of the coil. keep both inputs the same to keep current from flowing, pulse one or the other high to apply switching action. since they're only on for a few tens of ms, just long enough to make sure the switching action is completed, it doesn't fry anything
but if this is for the LED temperature safety switch, 60mA is nothing compared to how much power you're putting through the LEDs

[Anonymous No.1439567]

>>1439564
Not led.
I am making an universal overcurrent protection breaker which i can plug into any voltage and then set max current with a trimmer and then connect the load through the relay in the breaker. But since the breaker gets powered from the same source as the load i want to make sure it uses minimal amount of power, and keeping the relay on was the biggest power hog

[Anonymous No.1439570]

>>1439567
got it. one nice feature of the latching relay is that it can save state across power outages
another option that may work for you is to assume the common case is proper functioning, and therefore use a non-latching relay's NC contact to interrupt current to the load, so you're only drawing that 60mA once the load has drawn overcurrent

[Anonymous No.1439571]

>>1439570
and then the source voltage drops below the point where the relay can hold the contacts open and boom, house fire

[Anonymous No.1439573]

>>1439571
yeah, bad idea
go with the latching relay

[Anonymous No.1439585]

>>1439558
Even with no input voltage at all,
but not when you remove (!) XFG1.
Voodoo hidden in plain sight.

[Anonymous No.1439588]

>>1439585
C1 is backwards

[Anonymous No.1439595]

>>1439558
Q2's collector gets pulled straight to VCC-Vbe via Q6's BE diode. Any current drawn by Q2 also saturates Q6 -> Q6's collector sits at ~VCC -> Q10 buffers that -> Q8's collector is roughly at VCC-Vbe.

[Anonymous No.1439598]

>>1439595
Alright, makes sense, but how do I fix that? I guess I can put resistors in there to drop voltage to get it down to ((VCC-Vbe)+Vtail)/2 -> ((12-0.7)+1.25)/2 or about 6.3V but the whole point of this arrangement was to get the ridiculously high gain you get by using a current mirror active load instead of resistors. Loads of devices use this arrangement, granted, with a normal current mirror instead of one using Sziklai pairs. How do they avoid this issue?

[Anonymous No.1439602]

>>1439598
>How do they avoid this issue?
By not using Sziklai pairs. The basic current mirror uses local feedback to "regulate" the BE voltage, that is, the left transistor in your picture is capable of reducing its own base current by turning more on. Your Sziklai pair version can't do that. One of the reasons is that an emitter follower is not able to pull current from its emitter to VCC and the second one is that the emitter voltage of Q9 (on its own) would be Vbe(Q9) + Vce(Q5) below VCC, which is too much.

[Anonymous No.1439608]

YES! My buck babies from ping pong are finally here! No more lm317t degeneracy

[Anonymous No.1439609]

>>1439602
Well it's a bit better now. I appear to have a gain of about 288, which still seems low and the DC voltage on the output node is still 11.3V. If I put in a signal a bit over 1mVpk the top starts to clip, the bottom is fine for a bit until the output hits the 1.25V at the emitter. I really wish that output node would be at that 6.3V point so it could accommodate larger input swings. Increasing the tail current seems to increase the gain a bit, going from 150uA to 1.15mA brought the gain to 377 and dropped the output DC voltage to 11.2V. I could probably get it to 6.3V with enough tail current but it drops so slow it'd probably take over an 1A or something ridiculous like that.

[Anonymous No.1439610]

>>1439608
What are the specs of that dual heat-sink model?

[Anonymous No.1439619]

>>1439608
>not having a tracking preregulator before the lm317 to make it more efficient
>not having clean as fuck linear regulator output

[Anonymous No.1439663]

>>1439610
It can handle up to 8A up to 32v

[Anonymous No.1439664]

>>1439663
I seriously doubt those tiny-ass heatsinks can handle the 256W of power that unit allegedly puts out if you draw the full 8A at 32V.

[Anonymous No.1439665]

>>1439664
>I seriously doubt those tiny-ass heatsinks can handle the 256W of power that unit allegedly puts out if you draw the full 8A at 32V.

they don't dissipate all of the power. if the device had zero resistance, for a hypothetical example, it would dissipate zero watts.

[Anonymous No.1439667]

>>1439664
Uhh anon buck converter isn't a heater heat is only small byproduct

[Anonymous No.1439669]

>>1439665
>>1439667
I don't know what the fuck you morons are talking about. Listen, most SMPS are 80-90% efficient so whatever power you're getting on the output multiply that by 1.1-1.2 on the input. I'll be generous and say 1.1, so 90% efficient. 256*1.1 is about 282W of input power. That 282W is being dissipated by the switching element which in would be the transistors attached to those heatsinks. I don't suggest you ever try and draw the full 8A at any voltage greater than about 5V. You'll risk killing it.

[Anonymous No.1439671]

>>1439669
>That 282W is being dissipated by the switching element

my god, is this "I don't know english day"?

[Anonymous No.1439672]

>>1439669
That would literally break the laws of physics son

[Anonymous No.1439675]

>>1439672
No it doesn't. You don't get something for nothing. In the ideal case your output power would equal your input power however there are real losses to take into account so you end up drawing more power on your input than you get on your output. Your maximum output power is easily calculated. P = VI. 32V*8A = 256W. Where does it all get dumped?

<=====RIGHT FUCKING HERE

[Anonymous No.1439676]

>>1439669

hehe. 90% efficient means 90% is going to the load while 10% is lost in transit.

[Anonymous No.1439680]

>>1439675
I really hope you are just a troll, for your sake.

[Anonymous No.1439682]

>>1439609
>it's a bit better now
Not really, but the result seems to debunk the S pair theory.
Are you testing the component tolerance level of your sim game?

[Anonymous No.1439685]

>>1439680
I'm dead serious. What's wrong with you guys today? Usually /diy/ is smarter than this, holy fuck.

[Anonymous No.1439692]

>>1439680
I hope the opposite, sometimes you can't help being stupid. As a troll this would be pretty pathetic...

[Anonymous No.1439693]

>>1439685
>I'm dead serious.

you're either ignorant or trolling.

[Anonymous No.1439698]

>>1439693
Buddy, I've designed switched mode power supplies before, I know what I'm doing, you obviously do not.

Assuming an ideal converter with no losses, 8A*32V = 256W output power. Okay. In an ideal converter Pout = Pin, okay. So the input power is 256W. NOW, I don't know what range of input voltages this guy's particular ebay SMPS takes but something in the range of 12V is probably standard for a buck-boost topology like this so let's assume a 12V input. 256W/12V = 21.33A. Now, when boosting the voltage we know the FET gets shorted for a bit when the inductor is storing energy so you have the full 12V and 21.33A flowing through the FET. THAT'S 256W BEING DISSIPATED IN THE TRANSISTOR.

Do you understand now?

[Anonymous No.1439700]

>>1439698
I don't know shit about anything you just said, but it seems to me that if you've got 256W * 1.1 = 282W going into the converter and 256W going out of the converter, then the power dissipated in the converter itself is 282W - 256W = 26W. Is this not correct?

[Anonymous No.1439704]

>>1439700
>>1439698

He's been told that he does not understand what the word "dissipated" means, so I give up.

Have a nice day, guys.

[Anonymous No.1439709]

Can I use pic related to control a hot plate to keep a homemade smoker at temperature? I need to keep the enclosed box at 200 Farenheight. The hot plate is a standard 1000 watt Walmart deal, and will have a pan of wood chips sitting on top of it.

Link to thermostat: https://www.amazon.com/Inkbird-All-Purpose-Temperature-Controller-Fahrenheit/dp/B00OXPE8U6

[Anonymous No.1439729]

>>1439709
No it's shit. And overpriced. And big.
Buy $1.50 arduino from china, $0.10 thermistor and $1.00 relay and control the temperature from that. that way you can do even cool shit like add a motion sensor so it only heats the plate if there is someone in the room

[Anonymous No.1439730]

I want to buy some big ass LED that is like 5000 lumen or some shit, basically a sun in a can.
But i don't really know what to look for on ali express any ideas? they have many various leds there and it's confusing, i basically have 100W of power and 12V DC and want to get as much light out of it as possible

[Anonymous No.1439736]

>>1439609
Well, the left side of the current mirror is permanently one Vbe below the supply voltage. By symmetry, you can expect similar voltage on the other side. The base current unbalances the mirror a bit, though. If you want the right side to sit at 6V, you have to unbalance the circuit some way. In other words, it's not "meant" to produce huge symmetrical output voltage swings
And I suspect you'd get more (voltage) gain by replacing the remaining Sziklai pairs with normal transistors, because that would increase the effective output impedance.

[Anonymous No.1439742]

>>1439729
Is that really easier for someone that's entire programming experience is with ti-89 basic?

[Anonymous No.1439743]

>>1439730
You're looking for a blf q8. They're about $40 with coupon

[Anonymous No.1439744]

>>1439742
you can go to stack overflow and request a code or to arduino forum or reddit etc and then just copy paste it, it will be only like 10 lines or so so anyone will write it for you no problem

[Anonymous No.1439746]

>>1439743
>blf q8
Although it looks nice, you are paying for the casing and the other stuff, I only want the bare led

[Anonymous No.1439747]

>>1439744
If it's that simple i can probably learn it while waiting for it to arrive from China. Is it machine code or C?

[Anonymous No.1439749]

>>1439746
Shit somehow I missed the 12 volt part, sorry.

[Anonymous No.1439757]

>>1439747
>machine code
Arduino isn't the matrix son.
If it used machine code it sure as fuck wouldn't be 10 lines of code and nobody would use it.
Arduino uses it's own language similar to c++
But it is very simple.

Basically this is what you need to do,
the arduino has a function called loop, or something, i don't remember the shit exactly and i'm at work so i don't have the ide here, but there is where you put then code and then you simply read the input voltage drop across the thermistor, since arduino can't measure resistance directly and then calculate the temperature based on it
After that if the temperature is smaller than X you turn on a relay to power the plate or vice versa
If you wan't it even simpler, buy a digital thermistor they cost like two dollars and you will get the temperature directly as a number and won't have to deal with calculating the temperature from the voltage drop across the analog thermistor

[Anonymous No.1439763]

>>1439709

the only concern is if the relay output can handle 120V at 1000W. it doesnt say straight off, which is a bad sign, but there are 352 answered questions, so scroll thru those to see if anyone asked.

>>1439747

dont listen to the arduino zealot. even if you could put this together yourself, you'd need some way to calibrate it and linearize it. without calibration, it's completely useless.

[Anonymous No.1439764]

>>1439763
>you'd need some way to calibrate it and linearize it
Very cute. It's literally 5 minutes of work to make the whole thing. With a digital temp sensor you don't need to "calibrate" or "linearize" shit

[Anonymous No.1439766]

>>1439764

add the cost of a heavy duty relay, and your savings are zero.

and since you're a noob working with live voltages, you're putting your life on the line for zero savings.

[Anonymous No.1439768]

>>1439757
YOu can even go full on brainlet mode and use a raspberry pi zero ($10) instead of arduino.
The difference is that raspberry can run full fledged linux so you can write your code in any language you want, even Python (#1 language of choice of the programmers on the spectrum)

[Anonymous No.1439769]

>>1439766
>working with live voltages
OH GOD DON'T TOUCH THAT PIN IT HAS 3VOLTS AND 20ma IN IT
BZZZZZZZZZZZZZZZZZZZZZZZZT.. you just turned on a single LED

Yeah no.

The only dangerous part is connecting the heater plate to the relay. Just unplug the pleate first, screw in the two wires into the relay, then plug it in and dont touch it. I would let even my 10 yo daughter do it.

[Anonymous No.1439771]

Bought today some Ikea rechargeables just to see if they really compare to Eneloop Pros. According to some sources, both should come off the same production line. What surprised me is pic related, they recommend charging them at 0,1C which goes against all advice given for Ni-MH cells. You usually get told to go with 1-2C for proper delta peak termination. Guess they only sell dumb chargers under the Ikea brand that you need to terminate yourself?

[Anonymous No.1439773]

>>1439769

oh, yeah, reminds me you'd need to buy a metal case that must be grounded for safety, to keep the high voltage away from prying fingers. you'll spend an hour or two, drilling, filing, tapping, putting in grommets and stand-offs, etc. the cost of all this will likely exceed the cost of a completed thermostat.

[Anonymous No.1439780]

>>1439763
>the only concern is if the relay output can handle 120V at 1000W.
It says the relay is rated for 10 amps @ 250 volts

[Anonymous No.1439793]

If i want to make a usb plug for charging my phone, i need to only connect the ground and 5V right?

[Anonymous No.1439794]

>>1439780
Yes, but at 120V that is twice the current

[Anonymous No.1439804]

What's the cheapest lead acid battery I can buy or make?

[Anonymous No.1439805]

>>1439793
sure, its not like the data wires have anywhere to go

[Anonymous No.1439810]

>>1439780
>10 amps @ 250 volts

10A means it's good up to 1200W at 120V, so you're good.

[Anonymous No.1439811]

>>1439804
>cheapest lead acid battery I can buy or make?

go to kijiji or craigslist, look for ''auto parts and tires'' then reconditioned batteries.

[Anonymous No.1439844]

>>1438999
you don't need a smoothing capacitor for a welder it'll fuck up your entire current setting. just get the biggest bridge rectifier you can get your hands on and they can be rather expensive at that power level. fortunately if you're coming user of Google you can find manufacturers of large rectifiers and other electronic Components that with utilisation of a fake company name and purpose they will send you free samples. I'm not going to spoil the dish water by posting any examples of companies here but you know what to do. I have had literally tens of thousands of pounds worth of free samples over the last 25 years from some of the biggest companies. do make sure you give a good and realistic backstory to your company and product that you made up.

[Anonymous No.1439848]

hello anons poorfag here, I have electronics skills like soldering and other stuff-what kind of career should I look for?

[Anonymous No.1439849]

>>1439082
why would anybody want a DC welder anyway it's only handy 4 people building inverter machines. even on inverted machines they have to put a second inverted turn it back to AC or you can't weld aluminium. I say stick with AC box and if you want to add a tick tock then you only need to get the torch and a $1 HF start unit from AliExpress

[Anonymous No.1439851]

>>1439804
you will be tempted to buy cheap car batteries because they sell dirt cheap but they will be useless to you because you can't deep cycle them
so don't, buy an AGM battery

[Anonymous No.1439853]

>>1439283
tradition dictates you would start with a top octave generator chip followed by pre scalars. other ways to do it would be making a separate oscillator for each key from a few CMOS inverter chips. the modern way would be to use a microcontroller and some software to emulate the above. for example the digispark Clones you can get for less than $1 can be multiplexed to 16 keys. you can then use tables or pulse width modulation to get your tones. you can also utilise mathematical functions four different voices instead of having to use analog electronics. use Google to get schematics of old simple keyboards such as the vermona Hammond or stylophone

[Anonymous No.1439855]

>>1439283
as another side note you can also use a simple clock oscillator with a binary counter to drive the address bus on an old static RAM chip. these are junk box parts but they will allow you to record your own tones on each data line in square wave format

[Anonymous No.1439860]

>>1439848
car mechanic or plumber if you want to make any money. sub master's degree level electronics is dead service industry is dead. there is no way in hell that you can compete with the Chinese or Street shitters on prices of electronic modules which is why now everything is off the shelf and replace before repair

[Anonymous No.1439863]

can somebody explain to me why I would use a precision rectifier using two op-amps instead of first amplifying the signal and using a standard Bridge rectifier on the output.

[Anonymous No.1439876]

>>1439853
>top octave generator chip
Are those shits even made anymore?

>>1439863
Because the "precision rectifier" produces accurate results even at low input levels, while preamplifier + bridge just sucks less than plain bridge.
Also, a diode bridge needs floating differential input, which is usually inconvenient.

[Anonymous No.1439886]

>>1439805
>its not like the data wires have anywhere to go
some (many) phones use resistor voltage dividers on the data pins to set the charge rate

[Anonymous No.1439890]

very new to all of this:

I noticed this IC (one on the left) in my reel-to-reel tape machine is 14 pin, but is in a 16 pin socket and it seems like two of the legs of the socket were cut off to fit. Is this okay to do?

If so could I do this with the red sockets on the right? I want to replace them but only have 16 pin sockets.

[Anonymous No.1439892]

>>1439890
It's a hack, but it works.

[Anonymous No.1439893]

>>1439890
yes, it's fine

[Anonymous No.1439897]

>>1439892
>>1439893
Cool, thanks. I'm guessing it doesn't matter which legs I cut off? As in the two on the top end of the socket or the two on the bottom end?

[Anonymous No.1439901]

>>1439890
16 pin are more common therefor cheaper. so yes no problem unless it directly contacts the traces below and its high frequency causing some kind of inductance problem.

personally i only carry pin header strips and cut them to size because they fit. you can use a dab of hot melt/glue to stabilise the pin headers.

[Anonymous No.1439905]

>>1439897
it does but not electrically. one end will be keyed. this is to help you match the key mark on the IC so you insert the chip the right way around. its only cosmetic. what you have is only a plastic lump with straight wires running through it. there is no circuitry in a chip socket.

why do you need to replace the sockets? the point of sockets is so that you can just remove the chip. everything looks gold plated so you probably wont have any problems with the existing sockets.

[Anonymous No.1439916]

>>1439905
The red sockets are known for being really shitty and causing a lot of problems:
http://www.analogrules.com/MCIred.html

[Anonymous No.1439919]

>>1439916
ahh, fun. well i must also warn you that you will need lots of flux, a solder sucker and temperature controlled iron. also, use machine pin sockets because they hold the chip pins more securely than the spring type ones. gently break up the plastic. if you have a small drill bit or a dremel, use that to cut the socket in half and if possible cut the pins free very gently then make sure you have removed all the solder then use the soldering iron again to melt the tiny bit left and gently extract each pin. it's very easy to over heat vias and pull them out with the pins and damage traces. it's best to cut away the middle of the socket first and take pictures so you can repair any damaged traces later. sometimes there is absolute spaghetti under the sockets.

[Anonymous No.1439921]

>>1439919
Thanks for the advice. I actually already replaced 10 of them and I think they came out alright. Did a bunch of research beforehand so I've got the flux, solder wick, pump, ESD safe soldering iron, contact cleaner, all that. Got a few more I have to do on this board and another board.

[Anonymous No.1439930]

So i want to buy this cheap chink oscilloscope but it says the max input voltage os 5v, what the shit? That is useless

[Anonymous No.1439932]

>>1439535
>solid-state relay
>battery operated
Uh, solid-state relays are for high power devices, like the same as IGBT. Even an IGBT only uses power on turn on/off (and not 60mA). If you're looking for a triac, that would turn on/off AC, but battery operated AC switch sounds odd (I know the applications).

[Anonymous No.1439933]

Well /ohm/, I'm a computer's boy and really excellent in programming but I want to start to play with electronics stuff, I know that arduino is related with code but what other things can I programme? what languages use the PIC microcontrollers?

[Anonymous No.1439934]

>>1439680
It could easily be a gross mistake transferred from heatsink ratings/TDP and inverting efficiency. I made a similar brain fart the other day when I was considering a capacitor ripple: The power dissipation in the capacitor is not the sum of the area under the mean ripple*mean current, it involves the capacitors resistance.

[Anonymous No.1439944]

>>1439876
lol no, there are top-octave-generator boards these days

>>1439930
>useless
you are correct, sir

>>1439933
PIC micros use a 40-year-old assembler language and the programmer's model is uncomfortably constrained. most take 4 cycles per instruction. there are C compilers available. there are BASIC interpreters and there might even be a Forth environment available. but lolduino uses AVR, a much better performer, so your choice of environment is far wider (C, C++, Forth, the Arduino IDE, there might even be a microPython)
the arduino community is great on stepping complete noobs through making cookbook circuitry and getting it to work, but it isn't an extremely capable platform. you'll need to get into some electronics theory. if you know basic physics, The Art of Electronics (if it is available in your native language) is pretty good
as always, the question is, what do you want to do with it? if you have a project or three in mind, you will be much more motivated to understand how and why it works

[Anonymous No.1439963]

>>1439944
>what do you want to do with it?
I will like to open my curtains at 06:00, obviously make more complicated projects in the future but I want to start with this one

[Anonymous No.1439974]

>>1439963
arduino is probably the way to go. I'd be surprised if someone hasn't already made a recipe for this. alors, the web browser pioneer has done exactly this https://www.jwz.org/curtain/
since you already program, the mechanism to move the curtain is probably the most difficult part

[Anonymous No.1439990]

>>1439530
consider a latching relay. you just need that 20-60mA long enough to make it click into place.

[Anonymous No.1440020]

>>1439944
>40yo language
>use python my friend, it's very fresh
fuck that, C is the language of microprocessors. it is the only language you should be using to program them.

>>1439974
arduino is terrible. he probably just uses it as a PHY. they are notoriously reliable for crashing often. your best option would be to just crack open a cheap alarm clock, find a signal pin for the alarm(even easier if the alarm switches on an led for say the lcd display when the alarm is on for an easy to find signal contact) then tie that to a one shot 555 circuit that just sends a signal to a transistor to simulate the closing of a switch contact that you would find on a regular commercial curtain motor.

[Anonymous No.1440029]

>>1440020
40yo assembler languages are terrible. W register? srsly?
40yo macro assemblers with algebraic expression parsing, flow control, and memory management are patrician-tier
>commercial curtain motors
according to that article, they're shit

[Anonymous No.1440030]

>>1439944
>>1440020
>arduino is terrible.
Don't fall for the "Arduino is only for noobs" meme.

It's wrong. Arduino is fast, and fast is good, when you just want stuff to work.

[Anonymous No.1440042]

>>1439669
80% efficient means 80% of the power going to the load and the 20% being lost as heat in the mosfet. ie 256*0.2 is the heat dissipated by the mosfet
jesus don't do electronics ever

[Anonymous No.1440046]

>>1440029
>https://www.jwz.org/curtain/
no, according to the article they're noisy
so his instructions are to buy
>$100 for add-a-motors for salvaging the gears.
>$40 for arduino, ethernet and motor sheild
>$20 for weaker motor
>about $50 for the sound proof box set

>arduino motor sheild
>from the arduino official site
>You can also measure the motor current absorption of each motor, among other features
>literally comes stock with a feature to prevent stall/over current
>replaces the motor because he doesnt use the safety feature
>doesnt choose a stepper motor or servo
>easy as shit to turn a stepper motor back a few steps to unbind
>uses hall sensors instead of just sensing motor current to know when that shit is wound as far as it can go

I like JWZ. read his articles all the time and bought a few shirts when DNAlounge first opened. but with this he is being an internet of things retarded python fag.

>Arduino is fast, and fast is good, when you just want stuff to work
>what is stateless processing
you know what's even faster than a processor? a small amount of glue logic.

[Anonymous No.1440053]

>>1440030
if you're doing shit-tier projects, sure. eventually, you will outgrow it

>>1440046
>monitoring motor current instead of just sensing whether the damn curtains are open far enough
lad, I...
>stocking a dozen different types of DIP 74xx logic
drakeno.jpg
>stocking 10-20 each of those SOT-23 74xx1G57/58/97/98 chips
drakeyes.jpg

[Anonymous No.1440054]

>>1439698

This has to be an actual troll. But I just can't tell for sure.

[Anonymous No.1440057]

>>1440053
So unless you can go straight fromknowing nothing to knowing how to program the latest and greatest microcontrollers don't bother? okay nigger

[Anonymous No.1440060]

>>1440057
my point being, if you're coming from a programming background, don't expect arduino to offer you a multitasking OS. you're lucky to get malloc and enough memory to use it with

[Anonymous No.1440095]

>>1439930

all scopes are like that. they just add a big-ass voltage divider range switch to cut down the voltage arriving at the input stage. surely you can put together a voltage divider. it's like simplest circuit ever. even a 4chan diy'er should be able to handle that.

[Anonymous No.1440097]

>>1440095
I was under the impression that they used T or pi attenuators rather than basic voltage dividers

[Anonymous No.1440100]

>>1440097

dont confuse the noob, pls.

[Anonymous No.1440106]

Are there any tricks for finding proprietary circuit schematics? I feel like Ive tried every last trick in the book using google. We have a product pump down at work because of one of these fucking circuit boards and since they are discontinued, shit will be fucked for weeks. Since they are tossing the old card out, I figured I would keep it to mess around with. These things go for thousands of dollars and it would be ridiculous if it was a very simple fix.

Any ideas? Also, the board we have has similar white shit on the inductor coils. Does that mean they are burned out or bad?

thanks anons.

[Anonymous No.1440113]

>>1440106

the white goop is glue to protect against vibrations.

you wont find the info you want unless you have a connection inside the manufacturer's factory. you could search in elektrotanya.com but you wont find it. maybe put in a request for the service manual there, and pray.

[Anonymous No.1440135]

>>1440097
I can't imagine such attenuators will be good if you want to avoid frequency dependancies in your measurements.

>>1440106
It would only take 2-3 hours to trace that thing, provided you can see all part numbers. For an SMD board it's quite sparsely populated, and the traces are goodly visible.

[Anonymous No.1440144]

I want to run a raspberry pi off a few 18650s and a solar panel to recharge them, how do I figure out if the solar panel can recharge the batteries faster than the pi drains them (on average)

[Anonymous No.1440145]

>>1440144
Solar_power_solar*time_solar/time_total >= Pi_power_average

[Anonymous No.1440147]

I'm wondering if i should buy one of them current meters..
I have a multimeter but having to connect it in series every time i want to measure current is making me want to blow my brains out

[Anonymous No.1440148]

>>1440145
this will be my first electronics project I have no idea what this means - I'm guessing I need to figure out how many mah the panel delivers over 24 hours then how much the pi drains and compares but I have no idea how to do that

[Anonymous No.1440151]

>>1440147
so i am looking at the chink ones and msot of them say the measuring range is
>Measuring Current Range: AC current 20-400A
that cant be right, right? I mean if it only can measure current higher than 20A then it would be completely useless

[Anonymous No.1440152]

>>1440148
don't fuck with 18650s unless you know what you're doing.
>first electronics project
which, as kindly as possible, you don't. read the link at sticky FAQ #2, all of it. you should probably find a module that handles the charging and battery protection functionality for you, and just deal with feeding the pi from whatever voltage the batteries give you

[Anonymous No.1440153]

>>1440152
alright thanks pal

[Anonymous No.1440154]

>>1440151

low currents dont produce enough of a field to be easily measurable. there are units that have a 2A range but the price is considerably higher.
these things are useful for electrical guys, who handle big-ass machines, not electronics guys and their micro-amp circuits.

[Anonymous No.1440155]

>>1440154
So if i buy the 2A one is it only 2A and up? Because that is no good also. I need it to measure at minimum from 20mA and up

[Anonymous No.1440156]

>>1440151

low currents dont produce enough of a field to be easily measurable. there are units that have a 2A range but the price is considerably higher.
these things are useful for electrical guys, who handle big-ass machines, not electronics guys and their micro-amp abominations.

[Anonymous No.1440157]

>>1440155
There are specialty probes meant to measure small currents in PCB traces and such, but somehow I doubt you want to pay $hundreds or more for convenience.

[Anonymous No.1440164]

>>1440153
Energy in a cell (in Wh) is equal to capacity in Ah multiplied by average cell voltage, which I think is about 3.7 for a Li-ion cell. Convert it to J if you're an SI buff, but in general Wh is more useful. Figure out how much energy is needed to run the pi at the planned average usage rate for one day (power in W multiplied by time in h) and get cells for that. In reality you'll need less because the solar panels will charging the cells for a finite amount of time, but it gives you some leeway for inefficiencies and days with low sun. In fact I'd probably go for 1.2 or 1.5 times that energy capacity. Power/currents for usage of solar panels and pi computers can be found in their respective datasheets.

You may want to consider just running the pi of an appropriately sized USB power bank, as it eliminates the hassle of having to build an enclosure for the batteries and get a PCB. Probably not cheaper though.

[Anonymous No.1440174]

How do i make a jammer? Say, you have a router that is using the the 2.4GHz frequency.. how would i about hamming that shit?

[Anonymous No.1440176]

So usb2 plug has 4 wires and usb3 has like a million for some reason
But anyway, why the fuck are there two grounds? For what purpose?

[Anonymous No.1440178]

>>1440176
For high amp charging, which usb2 is incapable of.

[Anonymous No.1440179]

Anyone know of good lab power supply designs? After some months of heavy led blinking I think I am ready to try something bigger. I do not have enough experience to design my own and all I could find on the web were either hacked computer psus or "lmao just put 2 lm317t there bro". I have a 5A,12v smps here I am planning to use.

[Anonymous No.1440180]

>>1440176

https://en.wikipedia.org/wiki/USB_3.0

"First ask Google, then ask /diy/. Your question will probably be better received if you do so."

[Anonymous No.1440181]

>>1440179

if you can find some ancient LM723 chips, then you can get a current limited supply real easy. http://electronics-diy.com/30v-10a-variable-bench-power-supply.php

12V is not enough as a source, try 30V-40V.

[Anonymous No.1440191]

>>1440181
>12V
It is for 95% of hobby projects and for more just use a step up module

[Anonymous No.1440197]

I want to build a portable speaker, but I see no use in making it stereo.
I have a stereo driver but a single set of speakers, can I wire it up like in the picture?
If so: Will the speaker receive 50 Watt instead of 25 Watt?

[Anonymous No.1440199]

>>1440191

because the circuitry loses a few volts, you cant even charge a 12V battery (need about 14 for that), operate a 12V relay, or light a 12V LED strip adequately. etc, etc, etc. such a clueless comment.

[Anonymous No.1440205]

>>1440191

because the circuitry loses a few volts, you cant even charge a 12V battery (need about 14 for that), operate a 12V relay, or light a 12V LED strip adequately. etc, etc, etc.

such a clueless comment. there's a reason substantially every commercial supply gives you 30V+.

[Anonymous No.1440207]

>>1440197

no, dude. the two output stages will be fighting each other instead of helping each other.

if you need more power, look up how to make a bridge amplifier work.

[Anonymous No.1440219]

>>1440207
It's not about more power, but to get the total outcome of both speakers on one speaker. Like if the guitar is on the left channel and the drums only on the right channel and I want them to be on the same.
I know you can easily fix this with software, but I thought I could wire it up.
Thanks for your help!

[Anonymous No.1440221]

>>1440219

to do that, just mix the L and R signals and feed them into the L input. you just need 3 resistors. no software required.

[Anonymous No.1440224]

>>1440219

use a passive mixer to mix the L and R signals and feed them into one of the inputs. you just need 3 resistors.

[Anonymous No.1440230]

>>1440205
Not you dumb dumb, step up module gains voltage.
The only thing increased is the input current to satisfy the ohm's law and PSUs have plenty of amps to spare.

[Anonymous No.1440231]

>>1440230

in other words you're in favor of designing the supply badly, then fixing it by adding another module to address the defects. like, making a coffee cup in pottery class, and adding an optional handle later.

[Anonymous No.1440232]

>>1440224
That isn't a mixer, but a proportional adder. Mixer implies multiplication, which doesn't happen with your circuit. Either way he won't be able to stop the signal from feeding back one way or the other. If it's microcontroller-run he can maybe set it to mono, or otherwise send the same signal into both channels and hope nothing bad happens when he commons the outputs. I think he'd be fine running it on just one pair of outputs anyhow since it's one IC on one silicon die (assumption) so it wouldn't really make a difference whether he produced the output heat from one output or both.

Also those resistors would explode.

[Anonymous No.1440235]

>>1440232
>Also those resistors would explode.

they're on the input side, so they're getting maybe 1/10 of a volt.

>That isn't a mixer, but a proportional adder.

it's a mixer in the parlance of the audio world, tho prob not in the telecom world.

[Anonymous No.1440236]

>>1440231
If i already own an old coffee cup then it is cheaper to add a handle on it that is to buy an entirely new cup with a handle

[Anonymous No.1440237]

>>1440176
Wait a fucking minute!
I have been using usb3 ports on my pc for a long time and always plug my usb3 supporting devices into them, but since they are at the back of the pc i use usb extension cables. But since there are extra wires in usb3 that would mean i would need special usb3 cables and i have been using regular ones THIS ENTIRE FUCKING TIME FUCK

[Anonymous No.1440239]

>>1440236

having to fiddle with 2 pots on 2 different boards to do a voltage sweep is not worth the savings.

[Anonymous No.1440240]

>>1440097
Different scopes, different purposes, different solutions.
Bunch of voltage dividers + frequency compensation capacitors is a very common solution, particularly in older scopes. And your typical 10:1 probe is also built that way.
Pi and T attenuators are used when constant and controlled output impedance is needed, which isn't usually the case in scope attenuators.

[Anonymous No.1440245]

Do you think a peltier module can suck heat out of one of those cooling bags you bring to the beach?
I wonder if adding one would turn the bag into a small refrigerator.
I looked at the specs and the module is powered by 90W of power and that is a lot of fucking hear transferred

[Anonymous No.1440252]

>>1440245
Unless I misunderstood something, such things already exists. In other word, it is possible.

[Anonymous No.1440254]

>>1440252
Does it use peltier modules though? I don't want some shitty loud brick with a gas compressor in it and it also can't be some rocket technology which i can't buy for cheap and put in myself

[Anonymous No.1440259]

You're all nerds in this thread

NERDS

[Anonymous No.1440261]

>>1440254
Yes, they commonly use peltiers, particularly the cheapest models.
Before you spend time and money on this, maybe you should check what is available and at what price. And if you still want to make your own, you could check their power requirements, used insulation and such.

[Anonymous No.1440280]

>>1440181
How do you get 40v? Most transformers I'm finding are 24v max. I found a pc charger that is nearly 20v and 3 something amps, that can be used too I think. But the tension can be solved I think, I'm having problems with the circuit topology (I have no clue of where to start)

[Anonymous No.1440285]

>>1440176
the extra AARX/SSTX+/- pins are the 5/10gigabit connection, has 4 of them for a differential pair in each direction. GND_DRAIN is the shield for the high speed data, GND is the ground for power

[Anonymous No.1440289]

Kinda offtopic, but is it a good combo to study applied physics as major and electronics minor?

[Anonymous No.1440291]

God i hope i am not about to melt my expensive phone

[Anonymous No.1440294]

>>1440245
>90W of power
>assuming it is 100% heat transferred
It's not. You should do even the most basic research.

[Anonymous No.1440295]

>>1440291
what the fuck?
i set it to 5V and that is what the multimeter measured, but after connecting the load then output contacts measure 13V so i measured it again to make sure i'm not hallucinating and then it suddenly fell to 5V
I am scared

[Anonymous No.1440304]

>>1440289
>electronics
Do most places still offer them vs solely Electrical Engineering?

Applied Physics/Electrical Engineering makes sense if you want to do Photonics I guess? I mean you can do that solely within EE.

[Anonymous No.1440318]

>>1440144
RasPi 3 can consume a lot of power and complains a lot on the voltage, so it's not very compatible with a lot of things.

The Pi 2B is better on voltage and power.

For efficiency, they make the "zero" versions, but they're limited with the interfaces (makes sense though if you don't want hdmi for a solar panel project)

[Anonymous No.1440320]

>>1440295
This shit is getting wirder and weirded.
I tried connecting my kindle and it is only drawing 0.5A when charging (the battery is onlly 10% so its not like its full)
Yet when i measure current as a short circuit on the output it pumps out 3A which is correct, since that is the current limit i have set. So why is kindle pulling only 0.5A? There is no way that is by design since that would make the charging take forever

[Anonymous No.1440325]

>>1440320
>There is no way that is by design since that would make the charging take forever
It's by design.

[Anonymous No.1440328]

>>1440325
But why? Even shitty batteries can handle 1.5A easily

[Anonymous No.1440330]

>>1440280
>How do you get 40v?

a good place is old stereos or mini-stereos from the thrift store. for under $10, you can pull out a 25-100 watt transformer, along with the rectifiers and smoothing caps.

[Anonymous No.1440332]

>>1440320
0.5A is USB2(?) standard max current, I think higher current is negotiated?

[Anonymous No.1440335]

>>1440332
That would work in a way where the SOURCE limits the current by default and the charged device can ask it to allow more current.
But i have it plugged into the usb female as per the photo, with only ground and positive connected, and the converter is ready and willing to supply up to 3A if the device lets it.
It's the same way wall chargers work, the just show max current the can current into the device and let the device sort it out, and it's not like devices limit themselves to 500ma from a wall charger.

[Anonymous No.1440336]

>>1440335
*shove max current they can

[Anonymous No.1440340]

>>1440330
>how do you make a power supply
>find an old power supply

[Anonymous No.1440341]

>>1440330
Thanks, also is it normal to control the tension with a variable regulator? seems wasty. What are the usual aproaches to make cv /cc?

[Anonymous No.1440343]

>>1440335
Stackexchange and my multimeter say that shorting data + and data - advertises that the device is a dedicated charger, maybe the device doesn't attempt to pull more than 0.5A unless it thinks it's a dedicated charger? Do you have any power resistors or other loads to check out the voltage with high current draw?

[Anonymous No.1440344]

>>1440335
>and it's not like devices limit themselves to 500ma from a wall charger
The kindle does.

[Anonymous No.1440348]

So I watched a movie where they torture a guy by connecting car battery clamps to his balls.
And I remembered that i saw this in quite a few movies in various variations, like balls, titties, peepee, they all got grilled by the car battery.
So i got curious... I measured resistance of my ballsack with my multimeter and it's about 0.4Mohms. So that makes all those scenes bullshit. A fully charged car battery is about 13 - 14V. The current through your sack would thus be 0,0000325 amps and that is so small you wouldn't even feel that.
Like they have advisors and stuff when they make those movies, don't tell them none one of them knows the ohms law?
I even went as far as taking a small lithium 12V battery and placed the contacts on my sack just to really test it out... i felt nothing
What a joke has the cinematography become. They don't even try anymore.

[Anonymous No.1440350]

>>1440343
Fuck.
i already sealed the usb contacts with the shrinking tube.
So i have to short data + and -? Are you sure?
I tried to stick a 47ohm 5W resistor on the output and it did get warm, and while doing it i measured the voltage and it was stable 5V

[Anonymous No.1440352]

>>1440343
>>1440350
Tried shorting the data pins. Same result. 0.5A at most.

[Anonymous No.1440353]

>>1440352
Try the output voltage with a load smaller than 10 Ohms, the device will stop pulling current if the voltage drops too much (if the device is willing to pull more than 500mA to begin with)

[Anonymous No.1440356]

>>1440353
I don't understand what will that achieve?
When i measured current as a short circuit at the output i got 3A so the current is there

[Anonymous No.1440358]

>>1440352
Tried with my phone instead, only 0.32A this time ;_;

[Anonymous No.1440360]

>>1440356
The USB device will increase current draw until the voltage drops below the acceptable level, with short circuit I doubt the voltage was 5V, just put 5-10 47 ohm resistors in parallel if you have more of them and measure the voltage

[Anonymous No.1440364]

>>1440360
I tried with 0.1Ohms and it triggered short circuit protection so no dice there

Next i tried with a 10ohm resistor but it instantly burned my fingers so i had not time to take any measurements, i will hold a the multimeter probes in my mouth and try to be fast enough because it gets super hot in like 1second

[Anonymous No.1440365]

>>1440364
Fuck. Burned my fingers again. It gets hot so fucking fast.
But managed to take a measurement. The voltage never dips below 5V

[Anonymous No.1440418]

I dun goofed.
I have six nixie tubes being driven by binary decimal decoders.
So basically I have 2 tubes for seconds, 2 for minutes and 2 for hours.
Each decoder takes the binary input on 4 pins (takes it as half a byte).
So I have 3 shift register controlling the 6 decoders.
So there's a shift register feeding into the 2 decoders for seconds, one for the 2 minute decoders and 1 for the 2 hour decoders.
At the minute I'm just testing out my design with an Arduino while I learn PIC programming.
My basic plan was I could shift out the seconds, minutes and hours as a byte each and this would be passed to the decoders.
However as you can guess, shifting out 12 in binary B00001100, doesn't equate to the first second tube showing 1 and the second showing 2.
Each decoder is given half the byte so the first decoder gets 0000 and displays zero and the second decoder gets 1100, when the decoder sees a number greater than 9 it shuts the tube off.

In order for the tubes to show 12, I'd have to shift out 00010010 (aka 18)
So basically I can't simply count in binary, I need to count each nibble of the byte separately.
Is there an easy way to do this or am I gonna have to make a giant array?

Pic related, the shift register / decoder part of my schematic.

[Anonymous No.1440424]

>>1440341
>What are the usual aproaches to make cv /cc?

already answered up above >>1440181

>is it normal to control the tension with a variable regulator? seems wasty.

yes it's wasty, but it's the price to pay for beautifully clean power.
one trick is to use a transformer with multiple taps (10V, 20V, 30V, 40V) and have the supply switch taps automatically as you raise or lower the output voltage. this decreases the voltage across the output transistors, keeping them a lot cooler.

[Anonymous No.1440427]

>>1440418
either you are driving the shift registers with a microcontroller then you need to check every time it updates and do the necessary byte shuffling which actually isn't that difficult probably OR you can scrap the micro and do it all in logic chips, use e.g. a 74hc68 which is a decade counter, takes in a clock signal and gives you the binary from 0 to 10. that one in particular doesn't have a carry so you have to make one out of logic, a nor on all outputs or something? or you can have separate counters and control them individually from micro/shift registers. or find another counter that has a clock out.
the possibilities are endless.

[Anonymous No.1440429]

>>1440418
>Is there an easy way to do this or am I gonna have to make a giant array?

you just need 60 bytes to represent the numbers 00-59. that's hardly gigantic.

[Anonymous No.1440432]

>>1440418
>Is there an easy way to do this or am I gonna have to make a giant array?
output_pins = ((seconds/10)<<4) | (seconds%10))

[Anonymous No.1440451]

>>1438894
>mjlorton

[Anonymous No.1440454]

>>1439675
what an absolute retard

[Anonymous No.1440456]

>>1439698
>I've designed switched mode power supplies before, I know what I'm doing
>demonstrates complete lack of understanding of the subject
kek

[Anonymous No.1440518]

>>1440418
That's why I use counters that count in BCD, chained in parallel with the trigger output of one feeding the CLK of the other. Just have to emulate counting in BCD in the arduino.

[Anonymous No.1440520]

>>1440335
>and it's not like devices limit themselves to 500ma from a wall charger.
lol yes

>>1440427
>hurr durr use 74xx logic for everything
go back, 5V hipster

[Anonymous No.1440525]

Tell me three things i should put in my basket. Im ordering parts but cant think of anything cool.

[Anonymous No.1440527]

>>1440291
>>1440295
>>1440320

Your problem was answered here >>1439886
Your job is to find out what the device you want to charge looks for to decide how much current to ask for.
The device being charged controls the current being drawn by internal battery control circuitry.
Dumb USB with nothing on the data pins is only asked for 500ma.
Smart chargers tell the device how much it can supply by placing a small voltage on the data pins.
Different devices look for different voltage.
In some cases, the same device will accept different current levels depending on the dividers (1A, 1.5A. 2A, 3A)

[Anonymous No.1440528]

I'm still fucking around with diff amps seeing what interesting stuff I can build when I thought about a 3-input diff amp. Are there any useful applications for such a circuit? I'm lead to believe there aren't any since I can't find any reference to similar circuits online and even in testing it in the simulator it appeared to be unsuitable for amplifier applications and wasn't stable, even with feedback.

[Anonymous No.1440536]

>>1440525

If you've never tried these and you like hot sauce, try the original first and if you like it check out the others. It's not the cheapest hot sauce on the shelf, but it's not very expensive either.

[Anonymous No.1440538]

>>1440536
>posts only one thing
Thanks i guess

[Anonymous No.1440539]

>>1440538

These are good too. And for a third item, roasted and salted pumpkin seeds are amazing.

[Anonymous No.1440559]

>>1440528
ICs sometimes use multiple input and multiple differential input opamps internally. I don't remember seeing them with more than 2 outputs, though.

[Anonymous No.1440598]

Mechanical solder is for gay people

[Anonymous No.1440612]

Sup guys, I have problems with my remote control of my Blu-ray:

First, I thought the problem was the bateries, but nope, I did buy new ones and my control did still without work. Then I decided open up that shit to try to clean the circuit board as I did with another remote control(TV control) that now works 100%.

Well, I did it, but I think that I fucked it up: some little white pieces fall off from the circuit board. I tried to fix it putting tiny aluminium foil strips to complete the circuit. When I did the test to see if did works this was the result:
Nothing happens, but I got some asnwers only if I do squeeze the control like a whore squeezing a cock.

This means that I am on the right way, right?
How do I can fix that problem?
Do I should to use a Soldering iron to complete that white circuit instead?

>inb4 buy a new control you faggot.

[Anonymous No.1440616]

>>1440612
The white stuff is only ink, it's just to outline where things should be and to make labels. It's just printed on top of the circuit board.
The copper that the circuits are made of is underneath the green coating.

If it only works when you squeeze it, you should check out the battery contacts. make sure nothing is loose or dirty or corroded.

[Anonymous No.1440618]

>>1440348
Sweat might play a role, but yeah, 12 volts really is too low to be harmful. Having the cables clamped onto your balls is probably the only torturous part.
I think the hollywood myth came from using a car ignition coil to shock somebody. over time that got distilled down to just a car battery.

[Anonymous No.1440619]

>>1439730
Look through Cree's products. Find one you like. Then look for cheap knockoffs of that style.

[Anonymous No.1440640]

>>1440616
>If it only works when you squeeze it, you should check out the battery contacts. make sure nothing is loose or dirty or corroded.

What do I should use to do that?
Sandpaper or a tiny double-cut flat file?
Do I should desolder that?

[Anonymous No.1440656]

>>1439730
led street light

[Anonymous No.1440703]

>>1440527
This can't be it since many wall chargers just short the data pins and yet work fine with most devices

[Anonymous No.1440705]

>>1440536
this anon has excellent taste

>>1440703
>short the data pins
which identifies them as compliant to China's battery charger standard. apple does their charger ID differently yet
http://blog.curioussystem.com/2010/08/the-dirty-truth-about-usb-device-charging/
there are also charging control ICs that will attempt to extract maximum power without voltage droop from the input if it sees an ac adapter instead of a host
charger identification is indeed a mess, which is why USB-IF eventually promulgated a specification for it, and Qualcomm built their quick-charge spec atop that

[Anonymous No.1440717]

>>1440705
If that is the case then i don't see why my shit isn't working. My phone is chinkshit Xiaomi Redmi Note 4 and I shorted the data pins, so i am aligned with the chinese charging standard perfectly, yet it still only drains about .3A

[Anonymous No.1440718]

>>1440717
I will try the 200 ohm resistor after i get home to see if it makes any change but i doubt it and on top of that morale is getting pretty low already

[Anonymous No.1440719]

Is acid etching the simplest way to make a circuit board at home for cheap?

[Anonymous No.1440720]

>>1440719
etching has its attractions for complex or dense boards, but based russians just score it with a pen knife and lift it off

[Anonymous No.1440721]

>>1440718
Have a trimpot to jam in there?

[Anonymous No.1440723]

>>1440721
I don't need it to work with apple and other shit. I just want it to charge my phone and my kindle. So if the 200ohm resistors work for that since shorting it didn't that will be enough.

[Anonymous No.1440724]

>>1440723
I mean just for testing purposes, so you don't need to try a few dozen individual resistors.

[Anonymous No.1440725]

>>1440723
another link for your consideration
https://www.idt.com/about/press-room/xiaomi-selects-idt-wireless-charging-chipset-mi-mix-2s-smartphone-and-wireless-charging-pad
you may need to speak an actual protocol to unlock higher current

[Anonymous No.1440730]

>>1440724
Yeah i have lots of potentiometers and trimpots, i think 10k would be best for testing

[Anonymous No.1440731]

>>1440720
Holy shit scratching away all the copper you don't need with a knife would make me want to kill myself, not to mention the result will look like a shit
The best way is to print the circuit on paper and then iron it into the copper and then give it the old acid batch

[Anonymous No.1440732]

>>1440224
Thanks!

[Anonymous No.1440734]

>>1440731
The boards are usually just a bunch of copper with thin "anti-traces" in between.

[Anonymous No.1440735]

>>1440734
at that point why even bother and not just solder the component legs together directly?

[Anonymous No.1440737]

>>1440735
there is value in a substrate, I'm sure you're aware

[Anonymous No.1440742]

>>1440724
This is how a cheap car charger does it. 2V at pins 2 and 3 signals a 500mA charger. Chargers that use the 34063 instead need an external 75K+51K divider to generate the 2V from the 5V output. Some more info at https://learn.adafruit.com/minty-boost/icharging

[Anonymous No.1440744]

>>1440737
Pfft. Well, if you insist.

[Anonymous No.1440749]

>>1440744
I'm not gonna get mad at Manhattan builds

[Anonymous No.1440755]

>>1440742
Wrong quote?

[Anonymous No.1440761]

>>1440742
That seems to be entirely apple related article, but i only want to charge androids and it seems to be different with them
Is the charger schematic you posted for an android device?

[Anonymous No.1441458]

>>1440761
Yes.

[Anonymous No.1441461]

>>1441458
Then what equation do I use to derive the voltage for specific amperage?
If 2V = 0.5A ten 3A = 12V on data pins and that would be silly, so it must be calculated differently

[Anonymous No.1441468]

>>1441461
why aren't you reading the standard from USB-IF instead of pretending you know something?

[Anonymous No.1441469]

anyone has experience troubleshooting old scopes? I found one that turns on but the tube does not display anything.

[Anonymous No.1441472]

>>1441461
Since you won't pay attention to the other linked articles, you're welcome to ignore this one too.

https://www.eetimes.com/document.asp?doc_id=1279469

Amazon supplies a 1A charger and a 2A charger.
How does the Kindle know how much to draw from each one.
(I ended with a period because it's a rhetorical question)

[Anonymous No.1441474]

is there any other way to balance current in paralel power bjts than using ballast resistors?

[Anonymous No.1441475]

>>1441474
A current mirror? Kind of defeats the purpose but I guess it depends on what you're doing.

[Anonymous No.1441476]

>>1441475
they would be the pass transistors in a power supply

[Anonymous No.1441478]

>>1441476
other than tight thermal coupling between the transistors, emitter ballast is the easiest way in a power application. you'd be adding emitter resistance of some sort to measure emitter current for a balancing circuit anyway

[Anonymous No.1441480]

>>1441468
I read them and the problem is most of them says something different, float the pins, short the pins, put voltage on the pins, put resistance there .. etc... so far i know shorting them and floating them does nothing. I will experiment with putting a voltage on them next, even thought that seems to be mostly an apple thing, but whatever, i am frustrated and desperate at this point. I will try to find an old chink charger, rip it apart and measure the voltages and resistances they are using on the data pins

[Anonymous No.1441528]

>>1441480
>even thought that seems to be mostly an apple thing
I just measured the resistance of my factory Samsung Tab2 10.1 (Android) charger (without taking it apart) and it has 82k resistance between the shorted data pins and + supply and 82k resistance between the shorted data pins and - supply.
It won't charge while using it unless this supply or a similar one I made is used.
It will only charge at the slow/low rate unless this supply or a similar one is used.
The one I made has a 33k resistor to + and a 10k resistor to - and the data pins are shorted.
>>1441480
>I will try to find an old chink charger, rip it apart and measure the voltages and resistances they are using on the data pins
It won't matter unless it will charge what you're trying to charge at the rate you're expecting to get.
You don't have to tear it apart. Just stick a connector in it and measure the resistance between the connector pins.

[Anonymous No.1441530]

>>1441528
>It will only charge at the slow/low rate
*Even while turned off it will only charge at the slow rate

[Anonymous No.1441533]

>>1441480
>I read them and the problem is most of them says something different,
Because they are for different devices.
Early phones, later phones, small tablets, larger tablets.
>>1441480
>put voltage on the pins, put resistance there .. etc
Putting resistance between the pins is how the voltage is applied to the pins.
It would be much easier for you if you had some idea of how this eletrik thing works.

[Anonymous No.1441543]

>>1441533
>Putting resistance between the pins is how the voltage is applied to the pins.
No it's not. Putting voltage on the pins puts voltage on the pins. If you just put a resistor there it won't do nothing. Plus the usb standard says the data pins are supposed to be 0 - 200ohms, so the 200ohm advice was wrong anyway. Shorting them should be enough.

[Anonymous No.1441565]

I'm looking at ADSR circuits for my synthesizers

the ones I'm seeing need at least 10 volts on the gate/input for it to function
how do I step up a shitty little 555 synthesizer's output to 10 volts? An amp like the lm386n?

[Anonymous No.1441570]

>>1441565
Well if you're only making square waves then just use a FET, otherwise an amplifier of some sort is a good idea, provided you can get a nice central ground voltage. High-pass filters are our friends!

[Anonymous No.1441573]

What would I need to make a small Morse code transmitter?

[Anonymous No.1441574]

>>1441570
well the idea is EVERYTHING is going to be modular, from the waveform generator, to the filters, to the ADSR envelope

I'll have transistor-based triangle wave generators, 555 square wave generators, sine wave generators once I figure out how to make them, and ideally you could just chain a generator into the ADSR with TRS jacks or something

I don't know much about the underlying theory so I'm not sure why a FET is a better choice for pure square waves, why would that be?

I think I understand regarding the ground voltage, but probably not.
The high pass filter would just be to condition and clean noise from the signal, right? I've used them extensively in music production but not in electrical engineering applications

[Anonymous No.1441580]

>>1441480
Interesting. I tried connecting a Samsung 10 inch Galaxy A tablet.
With both data pins floating the tablet draws 0.4A and i get a pop up "slow charging mode, is your shit connected correctly?"
If i short the data pins then I get 0.62A and no warning about slow charge. But that can't be it. There is no way the tablet is content with 0.6A, if that was it, then charging would take forever.
So if it's happy to draw more current with shorted data pins, then why the fuck is it not drawing the full 3A or at least 1.5A?

On the other hand, the Redmi Note 4 won't even start charging at all with floating pins and draws 0.3A with shorted pins.

fucking hell
I am about to just give up and obediently slurp up the couple hundred mA of charging currents, which the devices have mercifully decided to accept, since it's still better than fcuking zero

[Anonymous No.1441581]

>>1441573
what would be the mode of transmission?

[Anonymous No.1441583]

>>1441574
A FET is by some measures an unregulated amplifier with a bunch of gain. Feed it a square wave, and it will sink current on and off on either side of its threshold voltage. Feed it a triangle wave and it's got so much gain you'll just get a square wave out again, though with an arbitrary duty cycle. Might want something a little more push-pull though even if you were just to use it for square waves, maybe a half-bridge. If you want to boost it to 10V with not much current output then an op-amp circuit is certainly what you're looking for, a pre-amp if you will. For driving any decent speakers you'll want something a little more specialised, for which I'd pick up one of those buck-a-piece Class-Ds.

>>1441573
Look at the pixie; not much.

[Anonymous No.1441587]

>>1441580
Apple and Sony are special (pic) and my noname 2.1A charger shows this:

1 5.12 V
2 (D-) 2.04 V
3 (D+) 2.74 V
4 GND

2=3 yields 2.43 V

[Anonymous No.1441601]

>>1441587
I understand why apple has it like that, all their stuff is retarded. But why are the cool companies like sony and Samsung doing it too?
I guess i could just make several resistor arrays and switch them with a switch to select sony, samsung etc, but it is pissing me off because it is a huge hassle for nothing.
The dedicated charger is also retarded, who even has such large resistors on hand? I sure as fuck don't. I will have to glue together like 10 small ones and the resistance is so high anyway that i might as well just touch the pins with my fingers as resistors

[Anonymous No.1441607]

>>1441601
I have a bunch of 1M, 2M, etc. up through about 10M. If you work with electronics regularly you really should have a large amount of every E24 series resistor between 10 ohm and 10M and a handful of smaller sub 1 ohm resistors for stuff like current shunts and a handful of power resistors, probably under 100 ohms is reasonable for higher power applications.

[Anonymous No.1441609]

>>1441583
Can't be pre made, I have to make it myself
>>1441581
I guess AM I've been told it's told it simpler

[Anonymous No.1441617]

>>1441583
>A FET is by some measures an unregulated amplifier with a bunch of gain

huh? I'm pretty sure FET transconductance tends to be lackluster and they actually tend to give lower gain compared to bipolars.

[Anonymous No.1441635]

I want to make an RGB LED light system that can project light with some sort of programmability to a wall, I'm wondering what kind of LED elements I should be looking at, high number of individual lights would obviously look better but are the lights strong enough? I guess I'm looking for a high-ish power LED strip that I can make into an array? I'll add lenses and shit for focusing

[Anonymous No.1441661]

>>1441635

google ''led light show'' or ''led light projector'' etc

[Anonymous No.1441671]

>>1440616
What do I should use to do that?
Sandpaper or a tiny double-cut flat file?
Do I should desolder that?

[Anonymous No.1441677]

>>1441574
Hey wouldn't it be great if you could derive all your different waveforms from one oscillator?

Well you can! You wanna start with a triangle wave oscillator, RC based relaxation oscillators are the go to (though I recommend replacing R with a current source to get nice linear ramps rather than that exponential RC time constant BS). In order to get the sine wave out you simply take the triangle wave and put it through a diode waveshaping circuit and now you have a sine wave. The square wave is even easier, simply put your sine or triangle wave (either will work) into a comparator based zero crossing detector and it will output a clean square wave with the same frequency of your input triangle wave.

If you decide you wanna do sawtooth waveforms it's as simple as setting a lower charge current to the charging current source and a higher current to the discharging current source or vice versa for fast charge slow discharge. This will also allow you to change your square wave duty cycle as well, however to do this you cannot put your square wave comparator after the sine wave diode waveshaping as the sine wave waveshaper needs a symmetrical triangle wave to work properly. Sine waves generated by sawtooth waves are weird and distorted but you may find you like the effect in your synth.

As for filters, I recommend active over passive. My preferred topology is MFB but it's really only suitable for low pass and bandpass filters. Sallen-key is probably a better choice as they are somewhat simpler to design and can do high pass as well. Butterworth and 0.1dB pass band ripple Chebyshev filters are suitable for audio and 2 or 4 poles should be adequate as well.

[Anonymous No.1441694]

>>1441671

dont use anything like sandpaper or a file coz it'll remove the plating, exposing the steel and leaving it vulnerable to rust. use an eraser; steel wool if that doesnt work, or a dremel with a brass brush if that doesnt work.

the brown stain is flux, and it makes no difference to anything.

dont desolder anything unless you suspect the solder joints are bad, in which case you can add some fresh new leaded solder to it to make 'em shiny.

i dunno what the original problem was, but, if the buttons are bad, wash the rubber keyboard with water and dish soap, using only your fingers to massage the soap in. the black lines on the PCB can be cleaned with alcohol on a q-tip. dont rub too hard coz the black comes off, and that's bad.

[Anonymous No.1441697]

>>1441671
>disc player
is this a joke?
was is thrust into the year 2000?
throw that piece of shit away and watch movies on your computer like normal people

[Anonymous No.1441743]

Brainlet here. I have two cheap AC desk fans I have been running basically non-stop for a year. They have gotten weaker and weaker and one got extremely hot, produced some faint hot electronic smells and stopped turning.

I lubricated the journal bearings where the axle contacts the housing and the stator and whatever in the actual motor. It does not turn, it makes a buzzing sound and gets very hot and freezes the axle the fan would spin on in place.

It does not look like the electrical motors I read about in books. The bearing on the axle is a solid metal donut, is not magnetic, with (now very shallow) diagonal serrations cut into it that are very worn. The outer ring has perhaps 16 or 24 small plates with all the copper windings on the outside. I do not understand how it works. There is very little play between the axle donut thing and the outer electromagnetic ring component thingie.

Any ideas on how to repair these devices? Both fans are 220v, maybe 30 watts, the kind you plug into the mains, both very similar but assembled slightly differently.

[Anonymous No.1441759]

>>1441743
All motors work the same way, current creates an electromagnetic field that reacts with a second magnetic field to produce torque between two parts.
The failure depending o n the type of motor will be a) brushes/commutator (applicable if the moving part (rotor) uses an electromagnetic field) are worn
b) insulation of a/some coils has deteriorated allowing a short circuit and so interfering with the creation of the necessary electromagnetic field.

By understanding the principle of a simple brushed motor it should be possible to reduce by inspection the operation of any motor configuration, simply consider how the two fields need to interact, what the source of electricity is, how the rotation will affect the field.

[Anonymous No.1441762]

>>1441743
>one got extremely hot, produced some faint hot electronic smells and stopped turning
Coatings on the windings wore out, windings shorted, and the windings got cooked. Can't fix that.

[Anonymous No.1441765]

>>1441759
My assumption is the rotor is worn to uselessness. If it was a brushed motor, it no longer has brushes at all so I'm assuming it was brushless motor.

Looking at it, it appears to be a 8 pole motor which I do't think is a thing. So I have no clue what type of motor it is or how it used to work or why it doesn't work anymore.

[Anonymous No.1441769]

is there a book for the theory behind all this stuff?
the OP has books for stuff like "How to arduino", "how to make X circuit"
but I want to learn shit like how the motors work like you all are talking about, how a sine wave gets generated and why, how a rectifier works, why you need to coil X amount of times on an inductor/transformer, so on and so forth

[Anonymous No.1441773]

>>1441769
>is there a book for the theory behind all this stuff?

It would be "books", and I like how you want to learn everything. That's a good goal when you are young and have time. If you can go to the book store at a good engineering university that has Electrical Engineering, look at the textbooks for first and second year, and look in them to see the level of detail. Things like motors are not covered well in university, in my experience.

Or, go to wikipedia and other sites on the internet and read a topic, and don't go on until you actually understand what it is saying. buy an oscilloscope and start building circuits and try to figure out why the subtle things happen, and how to read a spec sheet to see that those subtle things are sometimes described and sometimes they aren't. Look at manufacturer's sites for application notes; the better ones are crammed with information. Here's an example of one that you might spend weeks studying and still not learn everything that it can teach you: http://www.analog.com/media/en/technical-documentation/application-notes/an47fa.pdf

And this is just me, but I would steer clear of youtube. I don't think you can learn things like electronics or programming by watching videos.

[Anonymous No.1441778]

I'm trying to extend a USB webcam some 10m out into my room.
Seeing as tbat USB 2.0 can only achieve lengths of up to 5m before declining in signal fast I was thinking about those USb->ethernet extenders.
But there are no real electronics in them, so I was wondering if I coukd achieve the same result just soldering a adapter together in which the pins of the USB go to a couple pairs of a 10m Cat.5e cable.
Would this work or would I have interference or other hickups?

[Anonymous No.1441789]

>>1441543
>>Putting resistance between the pins is how the voltage is applied to the pins.
>No it's not.
>>1441587
>File: asd.png
Gee, look how they did that.
They put resistors between the 5v plus pin, the 5v negative pin and the data pins.
Presto magico - a voltage appears on the data pins.

A Little Knowledge Could Be a Dangerous Thing (especially when it's wrong)

[Anonymous No.1441794]

>>1441789
uhhh resistors aren't magic son, they drop some of the voltage and rest of it is applied to the pins, waiting to push through some current when a connection to a common ground is made

[Anonymous No.1441800]

>>1441769
for getting deep into motors you need a "electric machinery" book. I recommend chapman for starters. It has motors, transformers etc

[Anonymous No.1441801]

>>1441794
resistors are current/voltage converters

[Anonymous No.1441811]

Can I use this kind of circuit to charge a 9V battery with constant current? I don't have any rail-to-rail input op-amps, so I have to limit the current on the "low side" of the battery.

How should I stop the charging when the battery voltage reaches some limit, e.g. 9 V?

[Anonymous No.1441821]

>>1441778
There is nothing to say usb will magically stop working after 5m just that is not specified to work over 5m. In the same way it's not specced to work over cat5 cable.
Try connecting green/white usb to any color/white cat5 pair.

[Anonymous No.1441829]

>>1441811
You need to work on your 0.7V meme.

[Anonymous No.1441831]

>>1441697
I don't like watch movies on my computer. I prefer lie down on my bed as a Giant panda.

>>1441694
The original problem was that my control stop working. That's why I did thought that the problem were the batteries. I buyed new ones but nothing happened.

I tried that thing with the eraser. Didn't work, but seems that my the chances to get an answer by squeezing the control are higher than before.

[Anonymous No.1441848]

>>1441829
You are right... I just measured 369 mV at pin 3 and 196 mV at pin 5. However I do think the current doesn't need to be very precise in this application.

[Anonymous No.1441852]

>>1441831

if squeezing the remote fixes it sometimes, you likely have a (1) bad solder joint, most likely on the chip at the bottom, or secondly on the crystal, X1, or (2) a cracked trace on the PCB, or (3) the crystal itself could have a broken leg, which happens when the remote is dropped on the floor.
anyway, time to get an internet-programmable remote like the Logitech Harmony series. i now have 6 of them, dont ask why.

>>1441778

an active USB extension cord will go about 10 feet. to get 10m, you need a couple of boxes to convert USB to ethernet, and back again. these also need to be active (i.e. contain circuitry)

>>1441743

it's an induction motor, it sends electricity to the rotor thru magnetic induction. what you're describing is a classic case of lubrication turning into glue after a while. every cheap chink fan has this problem. you need to dump some grease on both bearings, and forcibly work the rotor back and forth, and round and round to get it loosened up. if you can get 100% free movement, you need to remove the rotor and clean the shaft. and make sure the bearings are perfectly perpendicular.

again, you do BOTH bearings, front and back. some lazy fucks just do the front coz it's easy to get at without opening the fan case.

[Anonymous No.1441862]

>>1441848
The entire circuit is overdone.
10K, 33Ω, BC548B x2, 9V1, done.
Standard charge 20mA 16 hours.

[Anonymous No.1441863]

>>1441831
>File: turtle with chips.jpg
Son, your turtle doesn't have a shell and it's gonna get 3 to 4 meters long and eat you.

[Anonymous No.1441867]

>>1441609
The pixie is (also) a circuit you can build at home, but morse is neither AM nor FM; it's CW.

>>1441617
Measuring voltage gain on either side of the threshold voltage and it's pretty impressive, regardless of what it is compared to a BJT. Only a mV or two on either side of V_Gth and it's already hit the rails on the output.

[Anonymous No.1441872]

How would i pick a mosfet for an electronic load?
None of them have a DC safe operating regime that goes up to the required voltage (30V) and with the desired current i wish to pass (up to 5A)
A lot of them have only 10msec or 1msec pulse duration regimes on the graph, and the ones that do have DC tend to only have like.. 0.1mA @30VDC
wat do

[Anonymous No.1441876]

>>1441528
>The one I made has a 33k resistor to + and a 10k resistor to - and the data pins are shorted.
I added two 100k potenciometers and set them to +/- 82K but so far no change
i only tried a kindle and a sony phone tho.
tommorow i will also try a samsung tablet and also changing the resistances values to the 10k and 33k

[Anonymous No.1441884]

>>1441821
well, there is the small detail of propagation time over the cable

>>1441872
here's just one possibility randomly selected from my collection of datasheets. it's a TO252 so a bit bigger on the board than the SOT23s you've apparently been looking at. the point is that you should look for low Rds(on) at the available gate voltage, which generally corresponds to higher Ids capability

[Anonymous No.1441909]

>>1441773
>>1441773
>Things like motors are not covered well in university, in my experience.
Nothing is covered well in university in my experience. Education is a scam for all but the 1% best schools.

>read a topic, and don't go on until you actually understand what it is saying
Sometimes you have to learn by doing. Ideally, you do both with reading slightly before doing.

>but I would steer clear of youtube. I don't think you can learn things like electronics or programming by watching videos.
Why not?

[Anonymous No.1441916]

I've been stuck on this snag for 3 weeks now.

I'm making a small device I solder onto a synthesizer that can control 5 pots via an arduino.

The 5 chips at the bottom are mcp41010 digital potentiometers. They're active-low SPI devices. The unconnected pins are Terminal2, Wiper and Terminal1.

The big chip on the top is a 74hc4051, an active-low multiplexer being used to activate the MCP41010s individually via their Chip Select pins. The 4051 is pretty straightforward, with A being the common pin routed to a0-a7, determined by the values passed to s0-s2, which are connected, even though i didnt draw it on the schematic im sorry.

The code is extremely simple: cycle through each combination of s0-s2 pins, each time sending a rising value over SPI, when it reaches 255, it starts over at 0.

The problem is, It only works for the furthest-right digipot. The rest of them do nothing.

Any ideas?

[Anonymous No.1441950]

>>1441916
those analog "switches" have Rds(on) of a few kilohms at 5V Vdd. quick fix: try increasing the resistors to 47k or 100k. proper fix: don't use pleb-tier analog muxes to generate digital signals

[Anonymous No.1441952]

>>1441916
>>1441950
>why
Vil on those digipots is probably around 30% of Vcc. the voltage divider you made with the analog switches may not be reaching close enough to zero to get there. try writing a test jig that just sets one "broken" digipot on CS and stops, but doesn't send anything. check the voltage at its ~CS and you'll probably see a voltage much higher than zero, confirming
>but what should you use
e.g. 74HC138 would be ideal

[Anonymous No.1441957]

>>1441884
Actually I've been looking at TO-220 and TO-247s
Besides if it was that simple I would have picked one out easily already

pic related, as per your selection

[Anonymous No.1441961]

>>1441957
Vds is going to be very low when the switch is on. Vds = Ids * Rds, so 5 * 0.01 = 0.05 V which is well within the dc SOA per figure 9

[Anonymous No.1441962]

>>1441957
It shows it can be used to pull 10A continuously with thermal resistance of 96°C/W, which is pretty high, not really heat-sink territory. No problem. Don't have to worry about those ms times at all unless you're going to be switching fast.

[Anonymous No.1441964]

>>1441961
>>1441962
But what about the case of if I have the mosfet fully switched on, and 30V is placed across it then? I'd have 30VDS across the mosfet, with up to the max set current (5A) on it, wouldn't I?
And it handles 10A continuously if VDS is like 0.15V

[Anonymous No.1441968]

>>1441964
Well then you're fucking stupid for using a transistor to short two power rails together.

[Anonymous No.1441970]

>>1441968
>presuming motive
maybe you're the one who is fucking stupid

[Anonymous No.1441972]

>>1441970
when the MOSFET is off, there will be no more than a few µA flowing through it. again, well within the dc SOA, by orders of magnitude
however, the pulse SOAs do suggest that you should turn the MOSFET on and off as sharply as you can by charging and discharging the gate quickly, to minimize the total time the MOSFET loses (not passes!) a significant amount of power

[Anonymous No.1441973]

>>1441970
If you have 30V across a MOSFET while it's turned on, then you are shorting something lest you have a load that's meant to be taking 3kA. If there's current limiting in the form of some sort of resistive load, then there won't be 30V across the MOSFET. It's as stupid a question as:
>what happens if I have a wire in a normal circuit and I put 30V across it?
It's simply not a situation you'd ever find yourself in. A MOSFET when on acts like a resistor, so the voltage across it will be proportional to the current being pulled by the load it is in series with. In practice, this load's resistance will be far larger than the ~10mΩ of the FET, hence the voltage across the load will be far larger than that across the FET, both of which adding up to 30V. Kirchoff's voltage law.

[Anonymous No.1441984]

>>1441973
It's an electronic load. You're saying that there won't be a case, intentional nor accidental, where 30V will be connected across the mosfet while it is fully switched on? Like for instance what if a large value capacitor is placed across it in some scenario? it'll definitely see a large pulse then, though yes that wouldn't be quite DC. Yes I know a power supply's output shouldn't have a large value capacitor on it, but i'm just trying to account for all potential cases
And that explanation is a lot more helpful than just
>you're a fucking idiot
Thank you for some patience

>>1441972
Yes I understand you and also what the SOA is getting at. But I figured there's some eventually where, intentional or accidental, 30V is placed across it being that it's supposed to be an electronic load for power supply testing.

[Anonymous No.1441993]

>>1441984
It's not customary to put a large capacitor across a FET, though you can see in that diagram that the thing has a single-pulse curve it can handle. Could probably extrapolate a total max energy dissipated into the FET from it, but you'd probably be fine with a 1µ cap or smaller. Any larger and you'd be stupid, but I can certainly imagine a 100nF noise suppression cap somewhere.

When designing circuits, it's fairly important to ensure that you don't get any shorts through transistor in any cases, as you may have seen a thread or two ago as one anon was trying to make a push-pull driver with his transistors in common-emitter configuration; when the driver input voltage was between 0.7 and Vcc-0.7 (or so) both the push and pull transistor would be on, creating a short between the rails.

In short, we don't design around faults by selecting parts that are resistant to any such fault, but rather with intelligent and rigorous circuit design. In case you're worried about burning an expensive FET in a freak event, you can put a fast-blow fuse in series with the power rail. In fact many home appliances use a fuse in conjunction with a capacitor or two such that if lightning causes too much of a voltage spike, the capacitor will die short and the fuse will blow. This stops the increased voltage from propagating through the circuit and frying expensive things like a phone plugged into the adapter.

[Anonymous No.1442001]

>>1441993
Ah that makes more sense.

Semi related, in buck and boost drivers, is the the voltage and current across and through the switching mosfet less severe because of the presence of the inductor then?

Are you an EE?

[Anonymous No.1442002]

Say I make a big (~1 cubic foot) capacitor by layering aluminium foil and candle wax inside a plastic container. The thing would store a fair bit of energy (up to around 700J) but the voltage would be really high unless I really decrease the plate spacing, which would make it too time consuming. So my question is how would I charge such a beast up to the full 200kV (for 25mm spacing)?

[Anonymous No.1442017]

>>1442002
https://electronics.stackexchange.com/questions/111585/how-can-i-construct-my-own-capacitor

[Anonymous No.1442024]

>>1441950
>>1441952
yep, that looks like it was the problem. The switch mechanics were never going low enough.
Ordered some 74HC138s, hopefully they arrive soon.

Thanks!

[Anonymous No.1442025]

>>1442001
I am distinctly not an engineer, just a physics student with too much time on his hands. In buck and boost converters, the inductors/flyback transformers do limit the current going through the transistor. An inductor is defined by V = L*dI/dt, maybe with a negative sign in there. Solving the differential equation for a unit-step pulse (V = 0V when t < 0s, and V = 1V when t > 0s) across the inductor and resistance of the transistor (and that of the inductor) gives something along the lines of I = V/R*(1 - exp(-t/T)), where T is the inductive time constant, equal to L/R. As the pulse is stopped after a short amount of time (1/(2f) for 50% duty cycle) the current can't reach anywhere near steady-state, V/R. The condition for this is that T = L/R > 1/(2f).

[Anonymous No.1442029]

>>1442017
I already know how to make a capacitor, I asked how to charge one with DC.

[Anonymous No.1442041]

>>1442029
connect it to a source of the desired volts

[Anonymous No.1442042]

>>1442029
You want two microwave oven transformers, the first will step up mains to about 2kV. Then you need some HV diodes and an HV capacitor both rated at least 4kV. 0.1-10uF for the cap. This rectifies the AC. The other microwave oven transformer's secondary will be connected from the 2kVDC rail to the anode of a high power thyratron like the CX 1536A. The cathode should connect to ground. The grid you can drive with a PWM SMPS controller IC or microcontroller. On the node where the transformer secondary meets the thyratron you want a high power rectifier tube that can handle greater than 200kV. Honestly you probably want it rated to 500kV to 1MV due to the large inductive spike. Anyway your capacitor goes on the output of this diode tube.

[Anonymous No.1442060]

>>1442041
Therein lies the problem.

>>1442042
Ah, so a big boost converter that runs at baseline 2kV? Sounds fucking awesome, but wouldn't the secondary coils need to be insulated to 200kV? That can jump 6cm through air, 2cm plus half a mm of enamel would be child's play, and that's assuming you don't have silicon steel to help you out. I think you'd need a secondary closer to a Tesla coil in shape; a solenoid that prioritises separation from either end of the coil. Because when the thyratron is shut off, you'll have the full 198kV across the coil. But isn't the inductive spike the thing we're trying to use to get up to 200kV in the first place? Even if the rectifier has some time delay on it the spark gap used to limit the maximum voltage to 200kV will prevent that.

Now since I think you'd need to wind a secondary anyway, would it be feasible to pair it with another coil as a flyback converter so you wouldn't need a HV thyratron, just the rectifier tube? Such a transistor/tube would be less expensive and can maybe switch faster, and I think the losses would be at least comparable.

If I'm correct, the turns ratio W of such a system would need to be such that the maximum voltage of the transistor/thyratron multiplied by W would give me my peak output voltage of 200kV, along with the margin for error. That's a lot of well-insulated turns, and with no magnetic core.

Thanks for the reply, done something similar yourself?

[Anonymous No.1442061]

>>1442060
Wait, you can't use a spark gap for feedback since it will just arc and discharge the capacitor (600J in a second or two would be pretty fun though), but I could just run feedback from the primary side to control the MOSFET. Using a MOSFET (or IGBT) to switch it also makes the whole thing far easier to drive with standard semiconductor components, what with the insulated gate and all.

Now that I think about it this is pretty similar to a tesla coil but without the resonance, but if I did make it resonant I think that would make feedback from the primary much more difficult to handle.

[Anonymous No.1442063]

>>1442025
Nice, I'll have to read up more on it. Thanks for that bit of math.
Likewise I was a physics student, but my electronics knowledge is quite lacking

[Anonymous No.1442070]

>>1442063
I've been taking specifically electronics physics papers that are somewhat outside of the normal physics curriculum. Not to mention spending a bunch of time here on /ohm/. Two years ago I was one of the kids here asking about how you wire LEDs, and now I'm solving the same questions for others.

[Anonymous No.1442120]

>>1442025
Not him, but i have a simple question, i feed 20V 1A into a buck converter, will I get 10V 2A on the other side?
As in I want to know if the buck converter will waste all the extra voltage its stepping down (like a say linear voltage regulator) or if it turns the extra volts into current instead.

[Anonymous No.1442123]

>>1442120
yes, minus a few percent efficiency loss in the buck itself

[Anonymous No.1442133]

Is it possible to split dynamic voltage?
I have a power source that varies between 24V and 13V.
I have two devices, one I want to have always 12V supplied to it, the other one i want to have whatever is left.
So if the source is 20V then device 1 will get 12V and device 2 will get 8V.
Since the source voltage changes i can't just a simple resistor voltage divider

[Anonymous No.1442161]

>>1442025
Switchers are dimensioned so that the inductor's current increases almost linearly. And thanks to saturation effects, the real waveform is about the opposite of what you'd expect from the L+R model, that is, it increases at slightly exponential rate.

[Anonymous No.1442162]

>>1442133
You'd need a 12V regulator which can both source and sink current. The simplest way to do this would be a zener diode regulator, but they're practical only at milliamp level currents. You'll get bit better results by buffering that 12V with an opamp.

[Anonymous No.1442169]

>>1442002
https://www.youtube.com/watch?v=gTt_YBzJ_Dk

[Anonymous No.1442170]

>>1442169
I did this and it worked, I used wood sealant because two part epoxy resin is fucking expensive in the third world. The sealant desintegrated after a while but it worked, about 7F at the start of the process.

[Anonymous No.1442173]

>>1442170
>>1442169
also I have no idea what tension rating mine was. I just stopped at 5v because the water would start to boil. You can also make some pretty hefty batteries with foil and these plates, but it's be good to use another electrolyte because if you use OH it'll eat aluminum (and release hydrogen) while the battery has no load.

[Anonymous No.1442175]

If I ever get thrown back in time to the medieval era, how do I make electricity to impress kaaaangz and shieeet?

[Anonymous No.1442176]

>>1442161
Doesn't core saturation point to a design flaw?

[Anonymous No.1442183]

>>1442161
I want to build my own switcher
That sounds like another problem
I've read one of the linear technology's datasheets and papers on building switchers and it basically just says select an inductor such that <formulas and a few explanations> and nothing about dimensions

[Anonymous No.1442187]

>>1442183
well, what about dimensions? the inductance and the max dc current (which covers saturation and heating) are the only real variables of interest at the circuit level, and both will be listed in the datasheet. just make sure your inductance is as <formulas> demand and max current is larger by a small fudge factor (20%?) than <formulas>
if you can't get an inductor that meets those requirements and fits in the space available, try a higher switching frequency

[Anonymous No.1442189]

>>1442183
>nothing about dimensions

you're not building furniture, dude. the fundamental characteristics of components are things like inductance, capacitance, resistance, Q-factor, current rating, noise current, etc. how those are characteristics are laid out in the eight dimensions of space is mostly irrelevant.

[Anonymous No.1442196]

>>1442169
This is cheaper and stores far more energy, and lets me mess around with massive electric fields. Well, cheaper provided I buy all my candles from opportunity shops for pennies. It's also orders of magnitude more deadly. Just imagine putting your hands across 200kV with the current carrying capability of 4-gauge aluminium wire!

For obvious reasons, I don't intend on attaching any kind of electrical load to this, unless I want it vaporised. It's also a long-term project that I'll likely not get around to this year, just throwing ideas at the wall.

[Anonymous No.1442203]

>>1442196
To be more safe, increase the resistance of your body by standing in the sun to create more dry insulating dead skin on your conductive meat

[Anonymous No.1442205]

>>1442189
Notions have a context. In electronics, dimensioning means to determine the values of components.

[Anonymous No.1442206]

>>1442205
which should have been covered as <formulas and a few explanations> to be figured in base SI units unless otherwise specified, no?
maybe a TI app note would be better for lerning2switcher

[Anonymous No.1442207]

>>1442205
Seconded. In physics, a dimension refers to any (independant) unit, as seen in the phrases "dimensional analysis" and "2N-dimensional phase space".

[Anonymous No.1442215]

>>1442205

reasonable people say what they mean explicitly, coz allusions lead to confusions.

[Anonymous No.1442217]

>>1442205
>>1442206

reasonable people say what they mean explicitly, coz allusions lead to confusions.

[Anonymous No.1442221]

>Switchers are dimensioned so that the inductor's current increases almost linearly
I'd say it's pretty clear that inductance and series resistance are the dimensions in question, along with magnetisation if you want to be picky.

[Anonymous No.1442264]

>>1442263>>1442263
>>1442263
>>1442263

[Anonymous No.1442270]

>>1442187
>>1442189
I'm not the one who brought up dimensions man

[Anonymous No.1442275]

>>1442176
Excessive saturation yes, but pic related is rather typical behavior for an inductor. The nominal saturation current for that part is 4A, which is based on 20% reduction in inductance.

[Anonymous No.1442583]

>>1441671
The solder joints on the battery contacts look good. whole thing look pretty good actually.
Clean the battery contacts with a brass wire brush. sandpaper only if you get desperate. it will take the plating off too.

When you put it back together, make sure the battery contacts aren't installed wrong into the plastic so they aren't contacting the batteries tightly.
If it works when you squeeze it, you have a contact problem, and the only place it can be is the battery contacts.
If the batteries are loose, stretch out the spring a little. not too much, dont break it.

[Anonymous No.1443435]

So every device runs on DC in the end right? So does that mean i can take say a TV, grab the traffo or the bridge rectifier inside, rip it out and replace it with a simple battery and it will work? Assuming i use a step up/down module to set a correct DC voltage?

[Anonymous No.1443479]

>>1443435
you're kinda right but not totally right.
Most things with switching power supplies will eat straight DC just as well as they will AC.
As a general rule, if the sticker on the power supply says something like "100-240v 5/60Hz" then it will run on DC too. no modification required.
Anything that has a transformer instead of a switching supply will blow up with DC, because a transformer is a dead short to DC.
Also, universal motors like you find in power tools and vacuum cleaners will run on DC as well as AC which is why they're called universal motors.

[Anonymous No.1443481]

>>1443479
>transformer
But if you disconnect the transformer and replace it's output wires with the wires to a battery then it should be fine.
I am not talking about connecting the power cable to the DC, but opening the device up, trace where it covnerts AC to DC, then remove that convertor thing whatever it is and connect a DC wires instead

[Anonymous No.1443484]

>>1443481
yes, you can do that for some things. If the transformer has multiple winding taps for different voltages, you'll need to have multiple batteries or regulators. Or if it has a center tap for positive and negative output voltage, you'll need two batteries.
If it uses the line frequency for timing (older TVs might? not sure) then you'll have a problem.

[Anonymous No.1443490]

>>1443484
What is the power lost on an inverter?
I wanted to just do battery -> buck converter -> device
But doing battery -> buck converter -> inverter -> device would be easire, but i fear that having to convert DC to AC just to have connected device donvert it back from AC to DC would be massively wasteful

[Anonymous No.1443586]

>>1443490
IIRC, inverters are 80% efficient on average. you're right, all that is wasteful.

[Anonymous No.1443594]

>>1443484
>you'll need two batteries.
to create negative voltage?

[Anonymous No.1443607]

>>1443594
that's the easy way. stack two batteries (or two battery packs) in series and connect the middle to ground.