/ohm/ - Electronics General

>>961059
I understand simple circuits but when they split like that on the left between the capacitor and the transistor then I have no clue what the person is trying to accomplish.
In this case my best guess would be a pull down bot the others I have no clue.

>Kerchov

>>961071

that's negative feedback. it shoots an inverted portion of the output signal back into the input to limit gain, and get stability.

>>961071
This is a simple darlington pair NPN BJT amplifier circuit. Each capacitor performs a slightly different function. The 10 uF coupled to the AC source is used to remove DC offsets.

The other 10uF one boosts amplification by acting as a less resistive path to ground for the amplified AC signal.

The last 100nF at the output is probably also used to block any DC offsetting of the amplified signal.

>>961059
For the pcb design add
Kicad it is an open source design program. It is funded by Cern and R-pi..
I have yet to use it as I have acces to altium designer.
But I have heard good stuff about it..

File: sad_rectifier.png (14KB, 301x306px) Image search: [Google]

14KB, 301x306px

>>961059
sloppy threadmaking

>>961417

That's a brit outlet config, right?

>>961519

nope, tube rectifier

File: 71GyoJGfgqL._SY355_.jpg (25KB, 355x355px) Image search: [Google]

25KB, 355x355px

I ordered a ESP8266 Devboard for $7

it'll be here soon, I am wondering why this is SO CHEAP compared to an Arduino to quote an instructable:

>>I found out that the ESP8266 is a powerful device by it self. And I asked myself. Why not use this device without using an arduino. This cheap (3 US$) device is more powerful than an Arduino UNO, got more memory (4000Kb versus 32Kb), higher clock speed (80Mhz versus vs 16Mhz) and 9 GPIO pins in version ESP-12

Hell the newer modules have more capabilities and they're still around $6 why is arduino stuff so expensive and what other cheap wifi modules are good for making IOT devices?

My project idea is basically just a button that posts a tweet every time I press it, or an LED that takes commands from a web domain, I shouldn't need to buy an Arduino Wifi shield for $40

>>961665
Chinaman is good at making cheap stuff. Also, the main chip in both products is cheap and has relatively little effect on the total price.

If you're happy with the Chinese chip/board, then use it. If not, buy better stuff and pay more.

>>961665
I've messed with it a bit a while back, just programming it straight from a cheap ch340 breakout.
the memory is severely limited and the code was not very easy for me to "get". might be better documented now.

i guess i stopped trying because i didn't really have a job for it.
internet synched clock maybe?

>>961679
>ardildo wifi shield
>"better stuff"

Fuck off, arduinobabby.

>>961665
Arduinos have always been marked-up toys for people with a passing interest in electronics. The underlying AVR microcontroller costs like $3 on Digikey.

That's not to say that the ESP8266 doesn't have drawbacks. GPIO pins are fairly limited compared to other general-purpose microcontrollers, and not all of its features are well-documented.

>>961683
Later versions are great - I have one running a weather display oled. The new esp32 is due out in april with a lot more features in. bluetooth 4, more and better adc etc.

>>961665
> I am wondering why this is SO CHEAP compared to an Arduino

Because Banzi is a greedy fuck.

>>961665
>IOT devices
A rope, a chair, and a closet would be more your speed.

File: 1433554483008.gif (2MB, 580x433px) Image search: [Google]

2MB, 580x433px

>inductor datasheet doesn't plot inductance over bias current

Are MOSFETS from ATX SMPS any different from general purpose except better high frequency tolerance? Does your everyday "I-solder-to-relax-and-make-blinking-lights-and-shit" guy have any use for negative voltage regulators? (fuck you Dell).

What are dual schottky rectifiers? I got a mbr3045pt or two.

File: 1436866694114.png (48KB, 475x242px) Image search: [Google]

48KB, 475x242px

>>961963
smps fets are the same as any other fet.

they're mostly high voltage and have a balance of low rdson and low charges/capacitances to minimize switching losses, so if you want to repurpose them to pwm a 12v motor at 1khz or something they're not optimal.

negative voltage is a meme.

the dual schottky is just two fast diodes in one package, to do pic related

>>961964

So If I used the meme by connecting gnd to positive and meme to negative, it'd work normally, yes? Any applications this might cause an issue in?

So the dual schottky works for center tapped transformers only? Pretty useless for me right now, then. Thanks for your help, mate.

>>961972
>using memes in your circuits

and the schottky can be anything. it could be two diodes in an interleaved boost converter. it can also be one diode if you short the anodes. normally parallel diodes is iffy but they're on the same chip so they'll share heat well.

>>961963
> Are MOSFETS from ATX SMPS any different from general purpose except better high frequency tolerance?
Nowadays, most MOSFETs are used for switching. It's just a case of trading off Vds[max], Rds[on], Vgs[thr], power dissipation, frequency, and cost. Choose the topology, work out the parameters, then pick the cheapest part which satisfies them.

> Does your everyday "I-solder-to-relax-and-make-blinking-lights-and-shit" guy have any use for negative voltage regulators? (fuck you Dell).
There are a fair number of circuits which, when built from discretes, are easier if you have symmetric power rails (i.e. you don't have to synthesise a logical "ground" that's half way between two rails). Class AB amplifiers are the most obvious example.

Hello /ohm/, anyone here had to explain how components work for equivalent of 10 year olds? I could improve my grade if I could figure a way to simply show how certain things work, like charging a capacitor and connecting it to a led with and without a resistor, or connecting a led to a DC motor to show it'll work when turning it only one way.

I need an illustrative and barebones method to show how they work. Anyone got more ideas? Theory and such I got covered, I just need attention-keeping experiments that don't involve sticking electrolytic caps in a mains socket.

>>962027

I wasn't asking because I'm building a SMPS or something - i desolder and solder stuff to relax and I picked some components from a very old Dell PSU, and wanted to know how I could use them later.

I read about negative voltages and such and they seem to be used mostly for stuff like op-amps, so thanks for an actually-useful application for later experimentation.

File: ternary-xor.png (25KB, 1008x879px) Image search: [Google]

25KB, 1008x879px

So i made a ternary (ie. not binary) NAND gate circuit, and it turns out you can construct other logic gates from NANDs just like with normal binary gates. I don't know if NAND is completely universal in ternary, but the ones I tested work. What are some more complex things that I can build entirely from NANDs?

File: ternary-nand.png (16KB, 642x642px) Image search: [Google]

16KB, 642x642px

>>962052
this is the circuit. (original content do not steal)

>>962052
You can build pretty much anything using NANDs. If you also use inverters and XORs, you can make your circuits extremely compact as well.

>>961082
Hey screw you buddy

I address him by a Russian name if I damn well want to.

>>962034
The methodology I described is what the SMPS' designers will have used.

SMPSes account for such a significant portion of the MOSFET market that "general purpose MOSFET" basically means something that's useful in a SMPS (with anything that wouldn't be useful there being a niche product).

>>962052
> I don't know if NAND is completely universal in ternary
Interesting question.

For binary, it's simple enough to brute-force it. There are 2^(2^2)=2^4=16 possible functions. Starting with the primitive functions {A,B,0,1}, you can compute the transitive closure of that set under each of the 16 possible functions in a reasonable amount of time.

The result is that 6 of the 16 functions are universal, corresponding to AND and OR gates with at least one of the two inputs inverted: !A&B, A&!B, !A&!B (NOR), !A+B, A+!B, !A+!B (NAND).

For ternary, there are 3^(3^2) = 3^9 = 19683 functions, and computing the transitive closure for just one of them may require testing up to (19683/2)^2 = ~100 million combinations.

>>961059
I made 2 guitar/bass pedals lately, both work really well and I fixed one of them today (one of the controls wasn't working) but the other one still has an issue. It's a simple volume potentiometer, signal comes in and middle contact goes out, the last pin is grounded. For some reason it's always at 100% volume though. I can just assume that the poti is dead at this point, I couldn't find any other potential problems with it. A picture wouldn't help since it's tiny and the wiring is messy, but yeah. I should really get a multimeter or a replacement potentiometer to check this.

>>962223
>the last pin is grounded.
Sounds like it actually isn't. Or maybe the pot itself is cracked near the ground end. You can check the latter by opening the pot.

>>962224
It definitely is, the solder joints are solid and it goes right on top of a ground plane. How do I open up a potentiometer? I never had to do that.

File: IMG_20160315_134623021.jpg (2MB, 3264x1836px) Image search: [Google]

2MB, 3264x1836px

>>962226
I think I might have broken it earlier, it's hard to tell in the pic but I think the trace in the top there is severed from the actual solder.

File: pd amp.png (39KB, 550x521px) Image search: [Google]

39KB, 550x521px

I don't get this op-amp circuit. There is obviously a current source (Ip) connected to the inverting input of the op-amp. But where exactly is it sinking its current?

Source here: https://www.maximintegrated.com/en/app-notes/index.mvp/id/5129

>>962249
current goes into the feedback path.

>>962249
Ground, reference voltage, whatever. See the actual circuit examples using photodiodes.

>>961059

> get 120vac 15a to 5vdc 1a ac adapter
> flip the transfomer
> ????
> profit

http://www.instructables.com/id/Make-an-Ultra-Simple-High-Voltage-Generator/

is this bait? how is he running a transformer with DC?

>>962305
>connect the transformer to the battery
>>one wimpy spark
>disconnect
>>one slightly less wimpy spark

Repeat as often as needed. Your "AC" here are the sudden changes in current when connecting/disconnecting the transformer. You also get flyback action on disconnect.

anybody know of a breakout board for LT's timerblox?

what is the best astable multivibrator you have ever used?

> inb4 555 timer, the "Integrated meme"

>>962306

interesting

so you could just do that with any old inductor?

File: s-l1600.jpg (125KB, 1600x1516px) Image search: [Google]

125KB, 1600x1516px

>>962308
Use pic related?
I liked the basic 2-transistor circuit as a kid, but I haven't really had that much use for astables lately.

>>962310
Any random large coil will produce high voltages when disconnected from a DC source, but if you attempt to replace the transformer of that Instructable with a coil, the spark energy will be immediately dissipated in your switch. So, you won't get long sparks.

>>962308
7414

File: IMG_20160315_195554_593490832.jpg (126KB, 500x498px) Image search: [Google]

126KB, 500x498px

Has anyone had experience with chinese "high voltage pulse generators"?
This one fmc800 is supposed to step up 3-6V up to 400kV, i believe these ratings are bs, but what could be the actual voltage with a 20mm gap?

File: graph.png (12KB, 439x420px) Image search: [Google]

12KB, 439x420px

>>962321

>>962321

I too have bought one of those before

they are still a total mystery to me

>>962327
Alright thanks, i've been searching for such a chart before.
>>962329
I guess there is a square wave signal being generated and then a transformer does the rest?
Anyways it has to be good as an igniter or for a rat trap.

How would I make a switch that is controlled electronically, and has a ten second delay?

I have a circuit that will output either 3.3V or 0V. I would like to make a circuit that detects 3.3V, waits ten seconds, then closes a switch. The switch should remain closed even is the 3.3V drops back to 0V, possibly requiring manual reset.

Is this easily achievable?

>>962336
You can use a relay for the switching and a 555 timer circuit for the delay, but then you will have to find a way to keep your high state at the relay.
Or you can use a cheap mcu for the delay if you have one around and code it so once it has an input it will output to the relay until you press a reset button.

Something like an attiny13 is a cheap and easy solution unless you don't know how to program them.

File: 9001h.png (13KB, 932x491px) Image search: [Google]

13KB, 932x491px

>>962336
This assumes that even a short pulse should trigger the circuit. If V+ is fixed, the diodes and one of the resistors can be omitted. "Load" can be a relay, too.
Reset is done by removing the supply voltage for >10 seconds.

>>962332
Flyback converter, similar to a car ignition. Allow a current to build up in a coil, then interrupt it.wind a secondary onto the same core to boost the voltage.

File: delay.png (6KB, 849x466px) Image search: [Google]

6KB, 849x466px

>>962336
Two S-R latches connected by a R-C timer.

File: Clipboard01.jpg (14KB, 412x404px) Image search: [Google]

14KB, 412x404px

hey /ohm/s,
quick question, can i replace the two C1 (220uF) electrolytic caps for virtual ground with something else, like 100uF? it's in the middle of the night here and that's pretty much all i have here.

Hi guys, i'm in an Internship/traineeship and i need to do a project for the company, what can i possibly do?

Thanks,

File: railsplit.png (5KB, 751x394px) Image search: [Google]

5KB, 751x394px

>>962418
Why try to synthesise a virtual ground when you can just connect to the point between the 9V batteries?

In any case: sure, you can use smaller caps, you'll just get more ripple.

Clearly, the current draw can't have any significant DC component (you'll only get 2mA through those 4.7K resistors).

If you have a spare op-amp, that will do a significantly better job.

>>962418
why is your ground virtual anyway? are you not really using 9v batteries? you could just connect ground to the middle of the batteries. the filtering caps are not strictly necessary but it depends on the circuit.

if the batteries are like an ac source or something then you can still do it with a centre tap transformer. if you don't have a centre tap then you need the virtual ground with the caps.

but to answer your question the size of the cap is proportional to the smoothing required. since there is nothing to smooth and no reason to smooth it you can fling both of them to fuck.

>>962441
>>962444
thanks, i'm just following a schematic, was pretty sure i could do virtual ground with just the resistors. and i just picked up a couple of 9v batteries i had lying around and they're not at the same voltage.
of all things, i'm breadboarding a headphone amp, the cmoy thing.
i'll try the op-amp virtual ground later.
thanks again.

>>962449
a virtual ground made of resistors the ground will vary as current draw changes.
It doesn't really matter if your +- voltages are different as long as there is enough headroom for your amp.

>>962444
> since there is nothing to smooth and no reason to smooth it you can fling both of them to fuck.
That's almost certainly wrong.

If the current drawn is DC, and the current draw is perfectly balanced (i.e. whatever is drawn from the +ve rail is returned via the -ve rail), then the resistors alone will be fine. But DC circuits rarely require split rails (and indeed he has since said it's for an audio amplifier).

An amplifier typically alternates between conducting between the +ve rail and ground and between ground and the -ve rail (i.e. the RMS ground current will be larger than that on the rails, as the ground carries current on both halves of the cycle while the rails carry negligible current half the time).

In that situation, if you ditch the caps then you'll either have to make the resistors *much* smaller (9V through 4.7K = ~2mA) meaning that you'll be sending a lot of current through them, wasting power, or the "ground" voltage is going to oscillate severely.

At 1kHz, a 220uF capacitor has an impedance of less than an ohm. But unlike a resistor, the impedance is purely imaginary, meaning that there's no power dissipation. Whereas a pair of 1-ohm resistors would conduct 9A and dissipate 162W (81W each).

>>962223
Just grab your test leads and check that the potentiometer is doing it's job across the sweep.

My last amplifier had a tone stack issue. Only the bass control was malfunctioning. Sometimes it would test correctly and sometimes it tested full pot value. I resoldered the three lugs. No change. I binned the pot and put in another new one. Problem solved.

It was a brand new pot, but the wiper must have been losing contact.

I ordered 10 each of red, blue, orange/yellow, white and green 5mm 4 pin "piranha" LED's off eBay and I'm rather disappointed in the light output of the orange/yellow and red. Running those at a higher current (30-40mA instead of 20mA) gives a light output closer to that of the blue and green but I'm not comfortable with doing that for most of their life. Is there a way to order LED's with more light output or do I just have to live with running them at a higher current? I bought these chinese ones just for lulz, at $0.99 for 10 I could care less if they an hero. Once I get everything working I plan to buy 100 LED's of each color from a reputable supplier so I'm cautious to run them a a higher current. Are my fears unfounded? More of CS than EE honestly so I'm not too sure about myself.

Tl;dr
>LED output sucks at rated current
>what do, double current or try different LED's

>>962618
The final answer would require the datasheets as different LEDs have different ratings, but generally speaking 40mA is above the absolute maximum ratings of run of the mill LEDs. Yeah, it's worded that way in the datasheets, so obviously you are not supposed to exceed it, even if it seems possible.
LED efficiencies vary widely and there's no reason to assume your cheap LEDs are particularly good. Since you got them from eBay, they might be even manufacturing rejects. Also, consider using LEDs with narrower viewing angle, as they look brighter.

>>962618
> Is there a way to order LED's with more light output
Yes. Check the rated power, or calculate it from the rated current and the voltage drop.

Better still, check the luminous flux (lumens) and/or luminous intensity (candela), as that will reflect any increase in brightness due to increased efficiency, whereas power ratings won't. Note that green LEDs will tend to have higher flux/intensity per watt than red or blue. A narrower beam angle results in a higher intensity for a given flux (intensity is flux per unit solid angle).

Voltage drop increases as wavelength decreases, so blue LEDs will consume (and emit) more power at a given current than red LEDs.

Don't expect to be able to run them in excess of their rated current. If they were capable of running reliably at 30mA the data sheet would say "30mA" rather than "20mA". Manufacturers don't downgrade their specifications just for lulz.

>>962636
>>962642
Thanks for the input! Is there a better/more reputable place to look for LED's? I looked at Mouser and Digikey but $1-3 per LED is rather steep, I'm leaning towards 1/3/5W LED "beads" and then just running the other colors at lower power to match that of the amber since amber seems to be the least powerful of the colors.

Sidenote, they'll all be run using PWM off an Arduino, does that mean I can use more current if my duty cycle allows them recovery time or will they just an hero anyway?

>>962636
>>962642
>>962964
As long as you can keep the temperature under control and avoid voltage breakdown you can dump as much current into an led as you like

>>962636
>>962642
>>962964

I found some of the LED beads on ebay, they say they're 3W. "Yellow (590-595nm): DC Forward Voltage: 2.2V~2.4V Forward Current: 600~700mA"

>>962964
Well those are going to be 50x as powerful as your previous LEDs. If you were looking for something on par with a 20mA blue LED, that's overkill by a large margin.

At that kind of power, you can use them for illumination (a good 5W LED lamp is equivalent to a 40W incandescent bulb), but thermal management (getting rid of the heat generated) will be a significant issue.

You also won't want to be losing a similar amount of power to a current-limiting resistor at that level (which will be the case if you're driving a 2.5V LED from a 5V supply through a resistor). High-power LEDs are normally driven via a switching regulator.

You probably won't find axial-lead bead LEDs above 50mA simply because that package sucks for heatsinking.

Farnell is showing 30mA LEDs for 10p and 75mA for 15p (single-unit prices). They have surcharges for small orders and minimum quantities for some items, but that gives you an idea of "baseline" prices.

>>962308
>>962312
>buys smd
>needs dedicated breakout board

ok in much the same way that i know how to move all the chess pieces, i know what the analog components such as transistors and capacitors do.

but i suck at chess, and suck at analog.

digital, however, is where im a viking. i can read a data sheet and talk to anything and make anything so far talk to anything else ive encountered.

i hate my job and want a better one. is there a position out there for a schmuck like me, and if so, what is it called?

>>962988
All I really want is LED's that have comparable brightness, blue being SUPERBRIGHTOMG and amber being "Lol is that all" doesn't do me any good. These seem to be closer matched based on the lumens listed on the page.

>>963022
In which case, go with the ones you already have and just reduce the current through the blue and green (use a higher-value resistor) to match the red/yellow.

If you want the same power dissipation from each, choose the current to be inversely proportional to the forward voltage.

If you want the same subjective "brightness", use potentiometers (or just chain fixed-value resistors) to adjust the brightness then measure the result.

File: SAM_0201.jpg (4MB, 4608x3456px) Image search: [Google]

4MB, 4608x3456px

who /flat bar toggle switches/ here

File: 14-pole rotary switches.jpg (102KB, 590x280px) Image search: [Google]

102KB, 590x280px

>>963378

sorry, but I only /round bar rotary switches/ big rough ones that can tickle my prostate.

>>963378
toggle switches are the best switches in my opinion witht that fucking satisfying clip sound

>>963378
I picked up a lot of these: http://www.ebay.com/itm/252007583418

They are amazing. Extremely loud, crisp click. Takes a satisfying amount of force to flip it, and it's also small enough to use on any project.

>>963529
I've got some Honeywell 6AT1s that look a bit similar to that. God I fucking love toggle switches.

how do i solve a grindy potentiometer

>>963047
I guess what I REALLY want is to make the amber and red as bright as the blue and green, as compared to dimming the blue and green to match the amber and red. White I have to dim for sure, almost blinded myself with it.

>>963703
You had those 600mA LEDs? There's no problem running some of them at 20mA and some others at 40mA.

>>963705
Those are ordered, the ones I have currently are the 20-30mA ones. I like the light output of the blue and green and want the amber and red to be that bright but my fear is if I run the amber and red too high they'll an hero and I don't want that to happen. Datasheet suggests 30mA but says they can handle 150mA for 1ms with an off time of 10ms but that's more than I care to do.

>>963712
That was explained many posts ago: your LEDs might die if you exceed their ratings. If you can't be arsed with pulsed drive, then you're stuck with the DC ratings. Which say no to 40mA.

>>963687

Grindy? As in it doesn't turn smoothly?

Does a solenoid's frame contribute at all to its "power" or strength?

I removed the coil from an open frame solenoid and it doesn't seem to propel the plunger as strongly as it used to. Then again, it could be the fact I no longer hear the noise when the plunger impacts the "cone" in the frame.

>>963712
> Those are ordered, the ones I have currently are the 20-30mA ones. I like the light output of the blue and green and want the amber and red to be that bright
Then buy some 50mA ones.

Also, check whether the issue is due to the beam angle. For any given power output, a narrower beam will make the LED appear brighter.

> Datasheet suggests 30mA but says they can handle 150mA for 1ms with an off time of 10ms but that's more than I care to do.
Apart from the added complexity, 150mA with a 1/11 duty cycle will result in the same brightness as 150/11 = 13.6mA DC, i.e. far less bright than just running them at 30 mA.

>>963735
> Does a solenoid's frame contribute at all to its "power" or strength?
Yes. See e.g. "Magnetic circuit" on wikipedia.

So I have this circuit, the guy says it runs 800MilliAmps for about 10 seconds but he has not given ANY values for the resistors, diodes, etc. used here.

Can someone help me out, first time project with a 555 timer IC.

http://www.lucidscience.com/projects/Pulsed%20LED%20Illuminator/4.jpg

I made this project with an Arduino but i want to make it with ICs to make it cost effective really.

I have a question for any radio whizzes in here that might have an answer for me...

So I understand that, due to electromagnetic voodoo and reasons, antennas smaller than about 1/4 wave suffer inevitable inefficiency, which may be acceptable for some applications and seriously detrimental for others. I also have ascertained that short dipoles and monopoles are linearly polarized, much like their larger counterparts; and that broadside-mode helical antennas are also linearly polarized, UNLIKE their larger, axial-mode counterparts.

So, my question is, presuming I had an application where size and polarization were critical and efficiency was not; would it be possible to have a very small antenna that is circularly polarized? Or is there some fundamental electromagnetic principle that precludes this?

>>963766
CMON man I got dubs, help me out here.

so what multimeters are good?

File: cockmeters.jpg (83KB, 800x465px) Image search: [Google]

83KB, 800x465px

>how do i solve a grindy potentiometer

the best solution is always replacement. if that's impractical, then if it's unsealed, you can spray contact cleaner into it, then twiddle it for 30 secs.
another option is to reverse the outside wires. this will make the pot work backwards, but it might avoid the bad spots - often a volume pot is never rotated past 50% coz too distorted.

>so what multimeters are good?

all of them.

File: Untitled.png (79KB, 1041x806px) Image search: [Google]

79KB, 1041x806px

>>963767
I know nothing about antennas, but Google says it is possible.

>>963766
R2 depends upon the voltage drop and maximum current of the LED (of which there should only be one; if you put LEDs in parallel, each one needs its own current-limiting resistor) and the supply voltage.

R3 depends upon R2 and the current gain (h[fe] or beta) of Q1. Once you've decided upon the collector current, the minimum gain determines the base current which determines R3.

R1, R4 and C1 determine the on and off times. See the 555 data sheet or a tutorial.

>>963732
yes

File: RCM schematic.jpg (22KB, 915x514px) Image search: [Google]

22KB, 915x514px

First time poster here. I have a record cleaning machine in the works. And I've got this 15 RPM geared motor that runs at 12v w/ stall torque current draw at ~600mA. What I want to know, is can a 12v walwart adapter be wired into the circuit? Is that safe? i was thinking of disassembling it and wiring it directly.

>>963842

yes it's safe. dont disassemble it coz then all the dangerous voltages become exposed. dont skimp on the current rating. try to get a 1A or better xformer.

>>963840

replace it

>>963926
but it's brand new

>>963980

Then it was shitty to begin with

>>963983
Yeah.
Lube and turning it back and forth for a while might help a bit, though.

Does anyone know if there's an IC for conditioning Thermistor resistance changes into a voltage range?

>>963835
Can you link me to examples?

My goal is to develop the ultimate FPV system for proximity flying in cluttered, confined places (i.e. through a forest or an abandoned building). There are currently two different approaches to this: one is to use CP antennas to cut down on multipath interference (which is undeniably helpful), and the other is to use a longer wavelength with better propagation characteristics through/around obstacles clutter (though usually with only a short monopole/dipole due to size constraints). I'd really like to combine both methods if possible.

File: DI5449f1.png (53KB, 600x531px) Image search: [Google]

53KB, 600x531px

>>964007
You need just a resistor. Or couple of resistor if you're trying to linearize the output.

>>964013
Like I said, I know nothing about antennas.
The search terms I used were "circular polarization electrically small antenna".

>>964025
Yeah, but the shitty tolerances kinda screw up the readings. I'm using a wheatstone bridge with an instrumentation amplifier.

Anyways, I found this:
http://www.ti.com/lit/ds/symlink/ina330.pdf

I could hook it to a 3.3Vdc source and read it with an Arduino Due. I think a 10kOhm based thermistor is linear over 0C-50C...right? Not completely sure.

My local homebrew computer club just shut down

:\

>>964033
Why?

>>964030
Use non-shit resistors then.
Bridges and instrumentation amplifiers are something you'd use with platinum sensors, which are relatively linear over the range 0-50C and have rather weak output. Normal (NTC) thermistors are exponential and their resistance changes roughly one order of a magnitude when temperature goes from 0 to 50C. There's very little point in using bridges and instrument amplifiers with them. The main problem is their nonlinearity, not the output signal strength.

If you actually happen to mean platinum sensors, then there are some conditioning ICs for them, like the oldish XTR101 & co. Many 24b ADCs also have all the required stuff for easy interfacing with Pt100/Pt1000 sensors.

>>964034
Not enough interest any more

>>964036
Ok. I still need an amplifier and bridges to span and offset the voltage output to a range of 0-3.3Vdc for the Due's analog pin. I'm not sure what kind of thermistor is, since it's inside an air handling unit and I'm trying to measure the air temperature inside it. I guess I'll have to get non-shit resistors.

>>964044
With an ordinary NTC you'd get around 1.6V output voltage change (using 3.3V supply) when temperature changes 0-50C, so you aren't going to win that much with your amplifiers. Personally, I wouldn't bother.
Even though I said "ordinary NTC" and "normal NTC", there are lots of them. There's also a chance that your sensor is actually a PTC, for example one of the silicon-based sensors. Your measurements aren't going to be accurate, unless you know the thermistor type exactly or go through the hassle of calibrating your stuff. In the latter case your initial accuracy means nothing, as long as your setup won't drift.

>>961059

do workers actually have to trim trim pots?

or do they just get a robot to trim them?

>>964061
If you want to eliminate the humans screwing your pots, you usually start by eliminating the pots themselves, instead of installing new robots.

>>964064

but low tolerance resistors are expensive

>>964068
Trim pots and trimming is expensive.
Software trims are often very cheap.

>>964068
They're fairly cheap when you go through a lot of them.

>>964074
Yup, it's all about the cal.

>>964068
If you're that cheap I'm certain you can just ask Vishay for some samples.

>>964084

my idea is to just set up an automatic trimmer so you can just get whatever resistor you need at very low tolerance, just for fun

>>964086
How is your idea different from digital potentiometers, DACs and laser trimming?

>>963839
What about them diodes though. I dont get why you'd put them in forward / reverse bias like that ??

I will read up on it. Thanks.

Reading up on how a 555 works should solve all these problems ??

>>964089

it's 500 times easier

How do I test if my rechargable batteries are good? Can I just use the capacitance setting on my multimeter?

I need terminology help, please

i have a chip that outputs i2s audio data as a slave. what kind of chip would act as a master and input? hope is to use an ft2232 with one channel controlling an i2c bus with stuff on it, and the other channel piping the audio data back over usb into the pc.

or am i thinking about this all wrong

>>964101
Abstract.

If I'm reading this right, you have a slave and you need something that acts as a master and a slave? i.e. a master to the slave, and a slave to the host computer? In that instance, it sounds like you're after a transceiver.

>>964114
i just dont know what to search for when it comes to i2s decoders because there's like 80 jillion data sheets to wade through. i2s->uart maybe?

>>963762
So keep the 30mA blue and green and just use 50mA red and amber LED's is what you're saying? Beam angle is like 120 degrees I think, found A datasheet but not the exact one for these LED's.

File: 10pcb.jpg (231KB, 3264x2448px) Image search: [Google]

231KB, 3264x2448px

I had one of my circuits made into a PCB about 2 weeks ago by Seeed Studio, one of those $10/10 PCB chinese manufacturers. I can confirm that you do in fact get all 10 PCBs for $10.

The only thing is that it could take a few weeks before you finally get your shit.

>>963839
See I need a switching time of 10microseconds, I need to push 800 mAmps through the transistor, and I have a 12 Volt input. I tried figuring it out but I cannot. Why would there be diodes ?? I know some guys use diodes for switching but I cannot put my head around this one.

Help me out here man. I am running only 3 LEDs right now, as shown - 3 green LEDs.

>>964152
SWEEEEET, what are these PCBs for ??

Why dont you get them flow soldered as well ??

>>964155
They were for a capacitor discharge circuit.

I didn't go for the whole manufacturing process since I already had the components on hand.

They have a $5/10 coupon going on right now.

>>964155
If I were him, I'd get PCB's fabricated and do all the assembly myself because it helps me relax and kill time.

That, and I love the smell of burning flux.

Gotta love those addictive carcinogens, y'know.

I do wish I had more excuses to use my soldering iron, as I have none.

>>964153
> See I need a switching time of 10microseconds,
The high time is ~0.7*R4*C1 and the low time is ~0.7*R1*C1. The period is the sum of those.

> I need to push 800 mAmps through the transistor, and I have a 12 Volt input.
Assume 2V for the LED (the actual figure depends upon the type, and could be anywhere between 1.5V and 4V), you have 10V across R2; 10V/0.8A = 12.5 ohms. R3 needs to be no more than R2 times the gain of Q1 (which could be anywhere between 10 and 1000).

> Why would there be diodes ??
It allows you to control the high and low times separately. Without diodes, the low time is determined by one resistor while the high time is determined by the sum of the two resistors, so the duty cycle has to be at least 50%.

> I am running only 3 LEDs right now, as shown - 3 green LEDs.
You can't reliably run LEDs in parallel like that. Whichever one has the lowest forward voltage will conduct most of the current. You need 3 separate resistors (which in this case would be 38 ohms each).

>>964130
> So keep the 30mA blue and green and just use 50mA red and amber LED's is what you're saying?
Yes.

>>964159
get some scrap electronics, you solder-loving fuck

Any electronics people do you know how to build a simple very low powered and short range mobile phone jammer?

There is a collegue at work who spends 60% of the time talking go his buddies on the phone, while at the same time pretending to do some work, management dont want to get involved because this guy is black (some people on the management are black too)

>>961953
You mean the current flowing through the inductor? If so then go for more well known brands like Coilcraft, Wurth electronics, or Bourns

File: cat staring at sandwitches.jpg (25KB, 363x321px) Image search: [Google]

25KB, 363x321px

>tfw needs to finish calculus ll to go into pratical advanced EE and Electronic classes
>tfw teacher is a romanian, who lived in chile that doesn't speak my native language.
Fuck my life. It seems it'll be Stewart, granville and khan academy for me this semester. Atleast I have Transducer laboratory classes which will force me to learn2arduinno/raspberry/PLC.
/endblog

>>961373
I only have good things to say about kicad.
The difficulty curve is smaller than EagleCAD's and also you have no limitations on it. Other than that they are on par.

>>964207
Scrap electronics are overpriced.

Free ones don't exist.

I live in ultra-Capitalism.

>>964279

> what is Goodwill

Can anyone explain why we ever use resistors in a circuit? All they do is get hot and consume energy.


Why are inductors and capacitors not used as a matter of rule?

>>964343

actually more resistance = less current = less power

>>964279
Go out the night before bulk trash pickup in your neighborhood and just drive around.

>>964343
resistors are generally used in logic or sensor circuits, not for power applications. dissipation doesn't matter much there.

some exceptions exist because resistors are cheap as fuck compared to their more efficient alternative. an example is a current sense resistor vs. a current transformer or a hall sensor.

>>964349
Not him, but all I see are those big-ass flatscreen CRT TVs and I can't really get them in my car

>>964357
Not having a beat em up truck you expect to die goddamn day. Were talking hundreds here.

>>964152
are they of decent quality or do the solder pads fall off as soon as they get hot?

I'm studying electricity and I have a kinda dumb question.

If I attach a wire between two terminals of a battery does this drain the battery more than if I attached the battery to a resistor like a heating element? Does it drain the battery at all?

Also, what happens exactly if one was to connect an AC directly to a battery? I imagine it ruins the battery, but why? Is it the negative portion of the sine wave pulling the 'wrong' way?

Also, could a full-wave rectifier/ inverter for two-phase AC just be two diodes or half-wave rectifiers in parallel or how does that work?

File: rectum.png (9KB, 542x229px) Image search: [Google]

9KB, 542x229px

>>964413
>If I attach a wire between two terminals of a battery
Lower resistance = higher current = battery discharges faster. The energy is dissipated (mainly) in battery's internal resistance.
>AC directly to a battery
One half cycle discharges battery, other ... does things depending on battery chemistry. You can expect it to partially charge the battery, though.
>full-wave rectifier/ inverter for two-phase AC
I'm not too sure what you're trying to say. Are you possibly referring to pic related? It was popular in tube era, not so much anymore.

>>964417
I guess it would depend on the battery chemistry.

Yeah, the circuit diagram like you posted, only backwards and with no induction coils. Two wires with AC waves offset by 180 degrees, each runs through a diode and then into a single wire with hopefully a full-wave DC current.

I'm trying to imagine an AC/DC inverter/rectifier without the use of induction.

>>964357
>>964368
Glad I have a pickup truck for this reason, CRT's are a good source of copper wire and miscellaneous components.

Just picked up some electrical contact cleaner from canadian tire, but now that I'm home I read that its for automotive use only, not for use on circuit boards and home electronics.

Is this going to be ok to use on my stereo or will it mess it up somehow?

>>964428

spray away, nigger. it's all the same: ethanol, hexane, isopropanol

>>964400
Different anon here, I've done SMT reflow with boards from this kind of service. Decent quality but the traces looked a little over-etched.

Alright fellow anons. I don't know if this is the right place to ask this kind of question but I'd appreciate if someone could help me. I need to build a music box, the one that plays a song whenever it opens. Building the box itself is not a problem but I have no idea what kind of eletronic should I use to store the music and trigger it when the box is opened. Could anyone help?

>>964463
I'd say a cassette player (because they have no boot-up time) rigged so when the box is opened, a switch connecting the power supply (battery) to the walkman closes the circuit and turns on the cassette player, same when the box is closed.

You can still buy micro cassette players for cheap. They're so cheap you can afford to buy two, gut one of them to see how they're wired and work from there.

Really the only thing you have to do is build the box and figure out the plunger switch.

Antique mechanical music boxes are different story.

>>964468
Alright, I'll look into it. Thanks!

So I am just getting started in electronics (had a couple courses of electronic & electricity in highschool), and I want to pick up a Art of Electronics. Should I buy a used copy of the 2nd edition for 25$, or save to buy a new copy for around 100$?

>>964491
buy a $0 version on google +.pdf

>>964185
T H A N K Y O U !!!!!

Really man, I'd buy you a beer. Thank you so much. I was going crazy reading astable mode, monostable mode and thinking which one would work and what not. You are a life saver man, I cant thank you enough.

So I can use any diodes I want ?? I bought a few for switching with the arduino - to get hardware input from switches and such, so it doesnt matter what diodes I use ?

>>964283
Doesn't exist.

>>964349
Trash disposal in my area is not publicly accessible.

>>964186
Well crap, now I wish I hadn't ordered those 1/3/5W Chinese LED's. Oh well.

>>964599
Which is? 'Merica here, never heard of it being done besides curbside pickup.

>>964644
>mosfet works differently, based on how deeply it is seated on the breadboard
It's just your breadboard being shit. Using a used mosfet isn't going to help and mosfet leads are generally too big for breadboards to begin with.
If you insist on using a breadboard, solder thinner wires to you fet before inserting it. Use different holes than you used earlier, as they might be fucked now.

File: nxuJh[3].png (7KB, 600x299px) Image search: [Google]

7KB, 600x299px

I've got a weird MOSFET problem.

If I push a mosfet in all the way to the bottom of the breadboard, the drain and source pins are always on.

If I only touch the breadboard holes, then the drain and source connection can be controlled like normal.

The MOSFET was originally soldered to this PCB >>964152 straight from its packaging. It was probably doing the always on thing while it was soldered since the capacitor could not charge (a previous board I ruined works exactly as expected until I short circuited the capacitor discharge output). The PCB is based on pic related + flyback diode.

Is it still working or did I get another broken MOSFET?

>>964400
Quality is decent. Just don't stab the soldering iron into the hole and you should be fine.

>>964645
I put a MOSFET I know to be working into those same holes all the way in and it works. That MOSFET has never been soldered to anything, only stuck into breadboards.

I want to drive a 3W LED with around 3.3V forward drop with a chip. Chip takes 3.3V as well.
I am thinking of using a 3.3V voltage regulator, that's being input with 9V.
And use a MOSFET as a switch for LED, that's being switched by chip.
So if chip turns the FET on, which drain is attached to 3.3V regulator, LED will drop 3.3V and take as much current it needs.
That's why I am thinking of using 9V before the voltage regulator, since both of them will drop 3.3V.
Will I burn something up? What sort of spec should I look for in the MOSFET?

File: IMG_1070.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

anyone have any idea what sort of input this is?

File: Midi_ports_and_cable.jpg (35KB, 780x534px) Image search: [Google]

35KB, 780x534px

>>964679
Looks like MIDI (or 5/180° DIN connector).

>>964679
It was a popular connector in audio equipment back in the day. Also used in MIDI. And since the connector was common, it was used for all kinds of other stuff as well.

>>964681
>>964682
Cheers. It's an old fire brigade transistor megaphone I found at a flea market today.

File: s-l225 (1).jpg (5KB, 225x119px) Image search: [Google]

5KB, 225x119px

>>964683

File: 10pc-Self-Locking-PUSH-BUTTON-SWITCH-Rectangle-Button-6-Pins-MOMENTARY-ON-OFF-DPDT-8-5x8.jpg (131KB, 800x800px) Image search: [Google]

131KB, 800x800px

Hey /ohm/, I've got a USB hub that has some buttons that turn different ports of the hub on or off. They look just like pic related. What I need to do is pop some of these buttons off, but I want to leave that port always on. I don't really understand these well enough to trust the pinouts I find googling variations "6 pin switch", so I was wondering if some kind anon could point me in the right direction

I was also wondering if there's a good way to embed components in place somewhere, so they can withstand stuff being plugged into them over a long time, and not eventually end up jostling around. But /also/ be easily removed some time down the line. The first thing that comes to mind, for me, is just getting creative with making a "hot-glue cocoon" of sorts

What I'm doing is putting a USB Hub, Bus Pirate, and RTL-SDR inside my Chromebook

>>964712
usb hubs don't cost enough that they're worth modifying in such a way.

>>964714
I could probably track down the same hub (shape roughly matters) without the buttons, and just order another one. Takes over a week to reach me though, and I am exceptionally impatient. Once I know where to shunt, it's not like it's gonna take long to do it anyway.

I think it's prolly worth it just to see how this thing works

>>964712
1 - switch port on
2 - desolder switch
3 - test continuity between pins
4 - once you've worked out the layout, solder wire where switch was

If the switch is laid out

1 2 3
4 5 6

Then it's likely that in one position it's

1=2 3
4=5 6

and the other

1 2=3
4 5=6

However it could be that 1 and 4 are the common pins, and the layout is

1 2 3
\=/

/=\
4 5 6

and

1 2 3
\==/

/==\
4 5 6

I don' know if this is the right place to post this question, but I didn't see a soldering general. I bought a naze32 rev6 acro with pins that I then soldered on....the wrong way. I desoldered them and have new pins, but the holes (don't know their technical term) I need to solder the pins into have solder in them. is there a good way to remove the old solder?

>>964712

Check the PCB tracks, probably only two are going into each button, so shunt those two. If not, use a multimeter to find out which pairs are closed when the switch is closed and shunt those.

>>964661

You still have to regulate the current. Good quality 9V batteries can output a lot of current - I test this by powering a 21W car bulb, if it lights up, the battery is damn good. The rest I don't understand, please try to be more coherent.

>>964434

It's just solvents? I thought they coat them with something to avoid future oxidization.

>>964916
Try putting new pins while holding the iron to the pads. Otherwise, melt it with soldering iron and thread stranded wire through it. Solder wick won't help you there, I think.

>>964924
Unfortunately the pins are all held together with plastic. I'll try the wire suggestion though, thank you.

>>964931

If you've got new pins, you can break them one by one and do the other method, too. It'll probably be unsightly, but it'll work

>>964583
> So I can use any diodes I want ?
If you don't mind the duty cycle being over 50%, you don't need any diodes.

Other than that, you just need to not exceed the breakdown voltage or maximum current, both of which are small, so just about any normal diode (not LED or zener) will work.

Also: D1 is redundant. Only D2 is required to allow a duty cycle of <50%. D2 effectively removes R1 from the circuit during the charging phase, so the charge (on) time is set by R4*C rather than (R4+R1)*C. The discharge time is set by R1*C either way.

>>964661
You can't drive a LED with a voltage source. Around the turn-on voltage, the current varies exponentially with applied voltage.

Process variation and temperature can affect the turn-on voltage. If it's 3.4V and you drive it with 3.3V, you get nothing. If it's 3.2V and you drive it with 3.3v, you get a dead LED.

So either find a LED which needs less than 3V, or add a 5v regulator for the LED.

>>961059
>tfw when like 12 problems of differential amplifier FET and BJT homework due in two days

>>964962
LEDs are not that ideal senpai. Most white LEDs can be driven significantly lower, and higher with color distortion/death. I was running a 30-34V "10W" LED with 24V, though it dropped all the way down to 20mA, instead the 600-700 rated.

>>964661
Roughly 99% of simple voltage regulators let you use a current or voltage setting. The current source you have to be careful with, because I^2 * Rset is the heat dissipation, and the Iout is usually Vref / Rset, so even with the lm317 putting out 1.25V as Vref, a 10 ohm 1/4 resistor would have to deal with .156W just for 125mA. If you got to 1A, you'd need very dissipative resistors. At that point, it'd probably be worth saturating a mosfet and using a heatsink.

>>965095
Edit, the LEDs must be more like 200-300mA. I was thinking of 9-10V ones.

File: 092809843.jpg (53KB, 690x157px) Image search: [Google]

53KB, 690x157px

Not sure if this belongs on /ohm/ or the stupid questions thread but I'm trying to go from 4 pole headphone jack and split it into two to go into the mic and aux jack of a motherboard.

What I have is 4 pole female jacks and 3 pole male jacks.

Stereo 3.5 jacks are:
Right signal
Left signal
Ground

Mics are:
Mic signal
Ground

So can I split it so the L + R + Ground head to the aux jack and the Mic + Ground (the same one) goes to the mic jack?

Headphones are Piston 3s.

File: 09820398403984.png (48KB, 731x529px) Image search: [Google]

48KB, 731x529px

>>965166
Thousand hours in ms paint

I bought a wavetek model 278 function generator a few days ago. Got it for a good price because it is having problems with the power switch i think. What happens is when i turn it on it just turns itself back off. If you hold the switch down it stays on, but when you let it go it will randomly shut off.

Yall think a wire is coming lose or the switch is going bad?

>>965229
No point in guessing. Open it and see.

File: IMG_20160321_140213.jpg (352KB, 2560x1920px) Image search: [Google]

352KB, 2560x1920px

>>964661
I draw a circuit that I have so far. It's more of specs that are needed.
Initial supply should come out of AC outlet.
One solution may be having an AC to DC 3.3V with enough amp rate, then add higher rated power resistor in series with LED. Seems like it would turn a lot of energy into heat.
Other solution may be having two different drivers for SoC and LED.

I am very open for some negative feedback since it's quite possible that I said something inaccurate. Thanks guys for your input so far.

>>965241
Like someone already said, driving a 3.3V led from 3.3V power source will produce somewhat random results due to tolerances and temperature variations.
N-mosfet won't work, unless the gate voltage is more than 3.3V. If the driving voltage is 0...3.3V, you need a P fet.

>>965096
>a 10 ohm 1/4 resistor would have to deal with .156W just for 125mA.
That would 156mW.
> If you got to 1A, you'd need very dissipative resistors.
1.25W.

Regulators designed specifically for current-mode operation typically have a lower Vref. E.g. the NCP3065 is described as "A Constant Current Switching Regulator for LEDs" and has Vref=235mV, so a 1/4W resistor would suffice at 1A, and 1/2W at 1.5A (the limit of the internal transistor).

But for a linear regulator, you can't get around the fact that dropping 9V to 3.3V at 1A means dissipating 5.7W somewhere (mostly in the regulator).

Is there any software that can validate circuits and give tips to improve them? I'm a fast learner and understand basic circuits, but I think it would help me a lot if I can get feedback on my circuits.

>>965251
There are digital simulators which compare your design against your test data and check shit like timing and then there are simulators (analog and digital) which try to predict how your circuit works. See OP pastebin for simulators.
Some simulators bitch about things which are kinda iffy and might suggest something else, but I haven't seen anything which would really focus on giving tips.

>>965169

it might work as is, but there may be a problem with the mic: usually the middle pin on the MIC jack provides a DC voltage to power electret mics. if your mic is electret, then it's fine as is, but if it's something else, then feeding it a DC voltage may silence it. if that's the case, disconnect the middle pin on the MIC plug.

small quibble: you keep referring to the earphone jack as the AUX jack. the AUX is an input.

For a lot of my projects i use AVRs, the one i use most frequently is the ATMEGA328P.
I can't decide if it is a dumb or a smart move to use that controller, for example when i try to use a common module or try to use sensors, all i will find on the web is stupid arduino code. I am at the point where i hate to hear arduino, i use the qfp package on my boards so i save a lot of space using an arduino is something i really try to avoid.

Is there an easy way to convert that code or even just use the arduino ide without the boards (would i have to add a crystal to it?).

I tried directly using the IDE to flash over a cheap ISP programmer that is not supported, it ended up with sync problems.
I tried getting the compiled hex file and burn it through the usual means, which worked but the mcu didn't do what it was supposed to.

How do you deal with that? I mean you seriously can't stuff an arduino into every project.

File: 1435754818645.png (252KB, 500x500px) Image search: [Google]

252KB, 500x500px

>>965477
there's a usb to serial programmer board for the arduino mini that you might be able to use with the arduino ide to program any 328p

https://www.sparkfun.com/products/9873
>inb4 sparkfun
shop around i'm sure you can find it for $4

>>965477
You hate arduino but don't have the knowledge to prototype a standalone MCU? Of course you need a XTAL if you're not using the internalOSC! Time to read up.

>>965480
I'll look into that thanks for the tipp.

>>965482
I usually used the internal oscillator and wrote all the code myself, now i try to use a display which is rather difficult to communicate with from scratch while there are libraries for it out there.
I was assuming that there are arduino boards that use the internal one, i didn't invest much time looking into these boards desu.

>>965492
Me again

https://www.arduino.cc/en/Tutorial/ArduinoToBreadboard
I just found this tutorial, this should work.

>>965169
>>965166
Unfortunately, 3.5mm jacks aren't completely standardized (this is especially true for 4-pole TRRS jacks, which have all kinds of permutations commonly seen). But for a TRS (3-pole) microphone plug, what you have illustrated will MOST LIKELY work, and the headphone plug is virtually certain to work.
http://www.hobby-hour.com/electronics/computer_microphone.php

>>965477
> would i have to add a crystal to it?
You'll probably need a crystal if you want to use USB. Full-speed (12 Mbps) USB requires the device clock to be accurate to 0.25%, and an on-chip RC oscillator is unlikely to meet that. Low-speed (1.5 Mbps) only needs 1.5%, which might be viable with the internal oscillator.

>>961679
>buy better stuff and pay more.

His question was how can it be better with those specs.

>>961665
can it do real time stuff? i hate the beaglebone because it lacks good real time shit

Question - Can i "read" the output from analog sensors by just routing the output (Op and GND) wires into a 30mm jack / headphones ??

>>965901
>30mm jack / headphones ??
holy hell that's some headphone jack

>>965907
You know you're a true audiophile when even the headphone plug is big enough to use as a dildo.

I have a 21 gauge thick full copper cable and I want to test 76mA LED bulb on it. Is this safe?

>>966035

stop being a nigger and google ''21awg wire current capacity.'' answer is about 5A since you're probably too dumb to figure it out anyway.

>>965901

sensors tend to have tiny signals, and are rarely in the audio range, so the answer is rarely.

>>966064
>stop being a nigger and google ''21awg wire current capacity.'' answer is about 5A since you're probably too dumb to figure it out anyway.

>21AWG
>5A
pick one

File: proof for the faggot.gif (23KB, 656x591px) Image search: [Google]

23KB, 656x591px

>>966099

Dont they teach sentence structure in grade school anymore? Or is it all meme shit now?

>>966119
Ok, my bad, thanks.
So it's good. I bought that 1W little LED from China, so just left with trying to solder them and fire the plug in.

>>965907
>>965921

All in or nothing boys.

Seriously though, the voltage goes from 0V to 5V - will it be enough to get a signal ??

>>961071
To be fair that's a pretty bad schematic drawing, lots of bad practices.

The cap on the left: Blocks the DC component of the input signal. The reason is that you don't want your BJT to be biased with a voltage you can't control. BJT's have an exponential relation between Base-Emitter voltage and collector current so it's not a thing you would want unless it's a controlled environment. The downside of doing this is that it puts a pole in your lower frequency band. Which means you are going to attenuate some portion of low frequency band so you should select it large enough to make sure that the frequencies that are of interest to you can pass.

The one at the emitter of the second transistor: This is a bypass capacitor, it basically bypasses the resistor which it's parallel to at frequencies of interest. The resistor is desirable because as I pointed out earlier the relation is exponential and putting a resistor in emitter flattens the relation making it more linear (also helps when BJTs start to heat up). But this is not something you would like for small signals because it reduces your gain, therefore you put a bypass capacitance to bypass that resistor for the frequency band you are interested in. You get best of both worlds by putting that.

The capacitor on the top right: This cap is the same as the first cap, instead of isolating DC of input it isolates the output. This could be used to prevent breaking the circuit you are driving. Most sound signals for example use 0 V DC bias so you wouldn't want to give DC bias to the output. This also has the same effect of preventing certain frequencies from passing.

To be honest this circuit is just something that is drawn for example at least I hope so because there are a great deal of issues with it so don't take it seriously.

>>966225
That voltage range is plenty for headphones or a speaker, but I'm not sure the sensor can deliver enough current at the signal to power a speaker/headphones without amplification. And even if it could, the signal would need to be modulated at an appropriate frequency (i.e. changing hundreds or thousands of times per second) in order for you to hear it. Putting a steady DC voltage to a speaker will not produce a tone.
https://www.youtube.com/watch?v=KaCWgQdo9wk

>>961665

Just get the Atmel Xplained Mini from Atmel

It has full debugging support and no tarduino IDE, so you can just use C

Also it costs 10 bucks (!)

>>964598

www.goodwill.org

If you live in a place with these

>>966326
Doesn't exist.

I want to etch my own PCB's but is it possible to use an other metal then copper? Can I for instance use iron or something else?

>>966467
Iron rusts and is kinda shitty to solder.
Aluminium is not solderable using normal solder.
Silver paint/paste is occasionally used, but this is about penny-pinching, right?

Tl;dr. Copper is the most practical material for normal PCBs.

>>966473
>tfw can't solder with aluminium
I already had this idea of etching with aluminium foil. Oh well, time to take out the buckaroos I guess

>>964261
yea im taking calc II for the second time and i'm still fucking it up. i spend hours and hours and hours working problems and I still do poorly on exams. Quizzes i do ok on though.

Hello /ohm/, can anyone remind me how are switches used to close a circuit when you open a door called? You know, either a pin or a ramp-kind-of-a-thing that opens the switch when the door is closed and closes the switch when the door is open.
I'll be installing LED lighting in my friend's cabinet and that's my preferred option for turning the light on and off, though I'm still thinking about reed switches. ebay links would be great, too!

>>966641
>remind me how are switches used to close a circuit when you open a door called?

they're called door switches.
http://www.ozautoelectrics.com/switches/door-switch.html

I'm interested in a low latency way of converting digital video signals to analog. What's a good way to approach this? FPGA and a DAC?

what is a good project to get into electronics?

I don't want some basic bullshit like lighting an LED

I want something a little more advanced

File: Triangle Osc.gif (4KB, 484x256px) Image search: [Google]

4KB, 484x256px

>>965250
Resistors are rated in watts, so it's better to describe the power delivered to it in watts.

>3.3V LED
And that's why you don't use them at 3.3V. I have 9-10V LEDs driven from a 12-15V input. If you had 3.3V, just put three in series (and be careful of shorts). Talking on DIY, a 60% efficient 15V->9V conversion is far easier and cheaper than an 80% efficient switch-mode supply. If you play closer to the drop-off, 12V->9V is 75% efficient.

>>966658
You mean like, VGA? A DAC LITERALLY is what you're trying to do.

>>966662
There's many kinds of electronics. For pure digital, the Raspberry Pi 3 has at least like 20 I/O ports, each of which can be oscillated at 200khz easy. PWM is even faster, but isn't usable for data. There are standard data channels, as well as USB. The included wifi is nice too, but the power draw is substantial now, you need a hub or 2+ amp usb.

Electrical Engineering is frequently signal oriented. I've literally been taking two semesters worth of amplifier classes, and before the midterm of this semester we covered differential amplifiers, like the kind in op-amps.

There's a huge gap between designing circuits according to electrical principles, and just using shit that should work.

If you want something more advanced, here you go, pic related.

The lm358n is suitable for such a purpose, being an 8-pin chip containing two op-amps. The frequency adjust resistor could be replaced with a potentiometer/varistor to adjust the frequency.

I have the DSO112 oscilloscope. It's ok for examining and just checking things, but without two channels and an easy computer interface it really has limited use. You can get two channel oscilloscopes and very high quality USB oscilloscopes for <$150. On Windows you're more likely to be stuck with whatever software they have, and on Linux you have to hope for support from existing software.

>>965245
>he also has the source at the high end of the LED instead of ground
He'd need like 13V on a common mosfet

>>964646
Not sure what you're describing with the mosfet, but mosfets can have the back metal plate connected to something internall. If it's touching the board, it would be forming a connection to one of the pins.

>>962193
>tfw one of my digital principles homeworks was literally making each two-input logic with only NAND gates
I think actually for ternary, you would have 3^27 possible gates. The 0-1-2 for inputs make for 27 possible input triads, and the output logic is 27 trits.
e.g the truth table goes from 000 to 222, for 27 combinations, and then the output logic states would correspond to 000000000,000000000,000000000 to 222222222,222222222,222222222


>>964033
My uni doesn't even have an electronics/hobbyist club.

>>964916
Desoldering braid exists. It can't fit into tight holes unless you cut it at an angle. It usually requires a very hot soldering iron to wick the solder properly.

>>966692
>You mean like, VGA? A DAC LITERALLY is what you're trying to do.

The digital signals are compressed so they would need some processing before simply being converted into an analog signal.

>>966703
> I think actually for ternary, you would have 3^27 possible gates. The 0-1-2 for inputs make for 27 possible input triads
I'm only considering 2-input gates, so 9 possible input pairs and thus 3^9 possible functions.

>>966723
Both DVI and HDMI transmit uncompressed PCM data. You still need some logic to handle framing, timing and decoding (the signals are transmitted using 8b/10b encoding), though.

I haven't really looked that far into it, but my intuition is that a CPLD would probably suffice.

>>961545
A sad rectifier

>>966238
But that is a static DC source - with no 'changes' so to speak. Hell man, I think I am gonna run 2 lines into a piezospeaker and see how that works out.

I mean i just need something to 'feel' the output of an analog sensor - a piezobuzzer is the way out lol

File: download (11).jpg (7KB, 230x219px) Image search: [Google]

7KB, 230x219px

Hello /diy/
Can one of you please recommend me a bench top power supply?
For use with Display drivers, Audio/visual Mosfet's ect, with oscillators, timing circuits and transistors
Don't need anything too beefy, just a /comfy/ unit

>>967226
you could convert an old ATX power supply

>>967229
I'm not sure if I clarified myself enough
After a variable powered unit,
IE 1-50V 0.1A -> 5A or whatever with banana plugs
Just need to power my breadboard circuits

>>967237
>1-50V
Literally none of those need 50V. If you're really after it, you can do +/-12 for an amp, but you could use a normal 12V rail and a 10-100mA negative rail from a SMPS.

old mosfets did need 10V+ to turn on fully, but trench and logic level ones can turn on at like 3.3 for 10 amps, or 5v for 100 or some shit.

The benefit of digital is the low power consumption and low voltage requirement.

For two of my EE classes which solely deal in amplifiers and shit, we use a 20 + 20 V .18A supply. .18 is the current limit and there are two rails, which make it easier to do a virtual +/- 20

File: parallelresistor.jpg (143KB, 2760x2449px) Image search: [Google]

143KB, 2760x2449px

Finally fixed the resistor catching fire problem on my solenoid capacitor discharge circuit.

I used this parallel jerry rigged 4 100 ohm resistors where one 25 ohm resistor used to be so that the wattage is distributed among all 4 of them.

where do you guys buy your wire? up until now i've been salvaging wire from various cables found at good will and bulk trash pickup in my neighborhood, but it's kind of a pain in the ass.

>>967261
Buy solderable magnet wire.

It is godsend.

>>961059
From afar that diagram looks like a roguelike. Would play.

>>967294
Circuit design based rogue like, when?

File: DSC_0164.jpg (2MB, 3200x3328px) Image search: [Google]

2MB, 3200x3328px

Hell/o hm/
Recently I got my hands on these old VQC10 (5*7 led matrix). I'm not 100% sure how to drive these lil beasts. I would like to make wrist watches. Only datasheets that I found were in germany (cause they were made in Deutsche Demokratische Republik). After translation I got some basic knowledge. So do you have any experience with displays silimilar to these VQ C10?
I can use AVR or Arduino or even LOGO! (jk, LOGO! is pile of shit).

>>967548
Based on the datasheet, it's a semi-smart display, which has the current limiting column drivers integrated with the LED matrix. So, you latch the column data for one row x 4 digits (5 bits parallel in, 4 character latch strobes), pulse the corresponding row input (you need quite a lot of current for that, btw), latch the column data for the second row and so on. You need to provide the character generator and the row drivers. All in all, bit tedious to use. The closest equivalent I know is Avago's HCMS-2700, but even it is quite different.

> I would like to make wrist watches.
Those displays are extremely power hungry, to the extent that you need to consider your battery's maximum allowed discharge current and internal resistance.

File: SATA power to USB.png (122KB, 1104x673px) Image search: [Google]

122KB, 1104x673px

Would this work? I have some spare parts and need a sata power to USB adapter, but don't want to pay the kikes at best buy $20 for one and don't want to wait a month for one to come from china.

>>967799
Oh, and this is for a 2.5" HDD, so AFAIK it should operate fine on 5V. Planning on running it off a 5V, 550mA USB wall wart.

>>967583
Thank you Anon. I really appreciate your effort. Well I will try to play with them on breadboard.
Wrist watches are maybe little bit to much right now. But when I saw them first I felt like they are meant for thing like this.
I have 3 of them so I will try make atleast IIC backpack for one so I can use at my "project". I would like to make locking system with numpad and display for my cabinet in school.

File: question.jpg (48KB, 1278x625px) Image search: [Google]

48KB, 1278x625px

Which type of device can do this in the endless sea of all kinds of NPN/PNP MOS-J-PMOS whatever FETs and transistors?

The current load would be 60mA.

>>967832
A PNP or P-Channel BJT/FET?

>>967835
You mean something well-known like a 3906 would do? Damn i though that this is harder.

File: talking-dice.png (283KB, 5060x3582px) Image search: [Google]

283KB, 5060x3582px

look at my shitty skills and cry.
Yeah, the switch is wired up wrong. yeah I only have 220 and 10K resistors. any way I can simplify the circuit?

>>962052
http://cnx.org/contents/vLMEHoQ0@6/An-algorithm-to-implement-a-bo

you should be able to convert any logic network into whatever logic network is more convenient for your BoM.

>>967871
1. What is the Arduino doing there? What function does it perform that you can't use the Pi for?

2. That's not how to wire a switch. As it stands, the input is left floating when the switch isn't pressed. There should be a 10k resistor between the input and V+ and the switch between the input and ground (so that the input is 0 when the switch is closed and 1 when it's open).

>>967908
As stated, I am aware that the switch is wired up incorrectly. It's already fixed.
The Pi was just the extension of my electronic dice I just slapped together with the arduino. Arduino and Pi are communicating, the Pi uses text to speech to announce the rolled number.

>>967911
> The Pi was just the extension of my electronic dice I just slapped together with the arduino. Arduino and Pi are communicating, the Pi uses text to speech to announce the rolled number.

Well if you want to simplify the circuit, the obvious approach is to ditch one of the processors and do everything on the same one. Using the Pi would probably be simpler, but the Arduino is perfectly capable of playing audio samples (given the small number of possibilities, there's no need to actually perform the speech synthesis in real-time).

Hey dudes, i have i problem, i know a little of electronics and i dont have any money to buy a better soundcard.
i'm using an behringer ucg102 on my guitar, the headphone output is on the line in of my x-fi, but i hear a lot of noise just by plugging on line in of my x-fi.. i dont know why.. any help?

(sorry for bad english)

>>968157
1. That concerns consumer electronics. Go back to /g/.
2. That is normal.

>>968182
Thanks for the info!

File: SAM_0220.jpg (263KB, 1152x864px) Image search: [Google]

263KB, 1152x864px

jesus fuck just kill me already

>>968343
>no flux
>multistrand wire
bro you just
>apply solder to the pads
>apply solder to the wire tips
>everything tinned
>now dip the tip of the wire in flux
>solder to board

>>968352
all I have is 1.6mm solder and a very thick conical tip in my iron. I have better shit coming in the mail, but until then it's 200 pound gorilla time.

https://www.youtube.com/watch?v=cvzhMYp8R1g
Jesus christ this guy is so fucking salty 100% of the time.

>>968354
>all I have is 1.6mm solder and a very thick conical tip in my iron

no excuses

File: SAM_0221.jpg (158KB, 1152x864px) Image search: [Google]

158KB, 1152x864px

>>968368
bruh it's like yarn

File: AYYLIEN.png (2MB, 1920x5696px) Image search: [Google]

2MB, 1920x5696px

Hey /diy/

Just finished my first circuit for PCB manufacturing, its a simple dual rail PSU. Can any layout pro's tell me if I made any simple mistakes? I did run a rule check in Altium but there were no errors.

Cheers m8's

>>967226
I got myself one of these: http://au.element14.com/tenma/72-2540/power-supply-1ch-30v-5a-prog/dp/2445412
Which is just a re-branded one of these: http://www.ebay.com/itm/KORAD-KA3005P-Programmable-Precision-Variable-Adjustable-30V-5A-DC-Linear-Po-/111455671404?hash=item19f346946c:g:vnQAAOSwKtVWy9qK

EEVblog did a review on them a while back and found that the old model had some big faults, but that's all been fixed with the new revision which is what you find on the market these days. As for mine, I have not had any problems with it so far and found that controlling it from the computer can be quite useful

>>967226
>>968380
Also avoid using a desktop PSU as a power supply as they have no way of limiting the output current. If you do go that route make sure to install fuses on all the rails.

File: LM337.jpg (89KB, 733x1026px) Image search: [Google]

89KB, 733x1026px

>>966662
Build a simple power supply, here is a good schematic to get you started.
This is what I started with for my power supply which you can see here >>968374

>>968373
I dont wanna teach eagle fly but if you steady hand and put solder just on the tip of iron you can get better results imo.

Btw what is it?

>>968374
I like that 3D version. Is there any way that I can do it with EAGLE?

>>968386
Not sure m8, I know that in Altium you usually have to create your own models or if you are lucky you can download a .step model from the supplier

>>968374
First mistake, DZ1/2 should be an 1N5355B which is an 18V zener. The one I was showing is just a normal diode.

>>968374
You might want to move the polyfuses before the regulators. That way they wouldn't ruin your load regulation.

>>968420
Are you worried about the resistance of the polyfuse's not being constant? From my understanding it's a pretty constant 1-2ohms up until you blow it

>>968454
I assume he's more concerned about the fact that the resistance is non-negligible, meaning that the actual output voltage will vary with the current drawn. Putting them before the regulators will result in the regulator compensating for any voltage drop.

>>963378
>>963449
>>963529
I disagree, every fucking toggle switch i ever touched will move it's contacts to the point of non-functioning when soldering because of the plastic melting before a proper connection is formed. It might be the quality of the switches I bought or my soldering skills. But except for toggle switches I'd say my soldering skill level is fine as I'm doing it for more than 10 years now. Did anyone else had this problem either?

>>968603
You must be doing it wrong.

It should take 600ms or so to make a solder joint.

>>968384
Sata data to ribbon cable to sata -> USB conversion board.

File: C1570P-00-2.jpg (36KB, 321x355px) Image search: [Google]

36KB, 321x355px

i want to buy a server power supply to convert into a high amp dc power source. currently looking at a power supply for a dell poweredge 6950 that is capable of 130 amps at 12 volts and only $50 new, but there are a bunch of different model numbers and i cant find an explanation for the differences. they are A1570P-00, PG237, NJ508 and U462D. does anyone know what the differences are?
also does anyone happen to know of a better power supply to use for about the same price?

>>968359
he only turns on the camera when he's pissed.

Its because he makes videos after he solves a hard/esoteric problem.

File: dual rail bipolar dual polarity regulated power supply using lm317 lm337 - output goes to zero.jpg (515KB, 1462x770px) Image search: [Google]

515KB, 1462x770px

>>968374

- if your transformer has a center tap, you can get more current out of it by using a full-wave rectifier

- it makes no sense to use the 2 op-amps to buffer the rails for displaying on the voltmeter. buffers are used on weak outputs when you dont want to load them down. however, the thing you're buffering is not weak; it's much stronger than the OP-amps supposedly helping it out.

- as others have said, the fuses should be placed before the regulator to improve regulation.

- the over-current indicator is less than impressive. it doesnt shut anything down, and it's not adjustable. an LM723 based regulator would be a better choice for proper current control.

- once again, if you have a center-tap transformer you can use a 2-diode trick to bias the ADJ pin 1.3V below ground so your output voltage goes all the way to zero. this is rather useful. see pic for example.

>>968721

look up each part number and pick the cheapest one. they're all meant to power the same server.

those rc car forums usually show how to mod them

>>968755
Yeah that's why I was looking at this one specifically. I tried that but nothing really showed up other than sales pages and they're all the exact same price

>>968751
- Transformer is not center tap, so can't use a fullwave rectifier or do the diode trick. Will have to remember that for future projects

- To display the -ve rail voltage on a cheap voltmeter I need to make the +ve, hence the inverting buffer amplifier. The only reason why I'm putting the +ve rail through a buffer is because I had an op-amp to spare so though why not.

- Can see concern with the fuses now, though if I move them to before the regulators won't that ruin the over-voltage protection given by the zener? As in if a over-voltage condition happens at the output (say I've accidently connected the output to 100V) there is no fuse to protect the circuit once the zener blows.

- I though the over-current indicator was kinda neat. Yes it does not shut anything down, it only turns on some LED's to show that the fuses (polyswitches) have blown. Was trying to keep things simple. Will have a look a the LM723 though, do you know if it has a brother for handling -ve voltage?

Cheers for the help too

>>968775
> if I move them to before the regulators won't that ruin the over-voltage protection given by the zener?
It's not worth screwing over the regulation for the sake of protecting against that scenario.

Over-current protection should be handled by the regulator. If you want to uses zeners to protect against connection to a voltage source, use a regular fuse.

A more fundamental issue is that if the current drawn from the +ve and -ve outputs isn't balanced, the current drawn from the transformer will contain a DC component. If the DC current is significant, and the transformer doesn't have an air-gapped core, it will saturate the transformer core, either blowing the fuse or (if there isn't one) the transformer.

Check the data sheet of the transformer to see how much you need to de-rate it for a given DC current. Or just scrap that design for one which doesn't use half-wave rectification.

Actually, just scrap it and use switching converters instead.

File: fds.png (8KB, 811x415px) Image search: [Google]

8KB, 811x415px

Record cleaning machine guy here again. I have a 16/3 10ft extension cord I plan to use as the power cord. And I want to chop out ~1ft from it to use for the vacuum motor, which will be plugged into the power cord outlet. My question is, what should I use to reconnect the power cord. It'll be cut out 4-5" from the outlet assembly. Thought about just soldering the wires and using shrink tubing. Or perhaps using crimp connectors.

File: images.jpg (6KB, 225x225px) Image search: [Google]

6KB, 225x225px

new to electronics and owning a home. i was trying to pull an old breaker box from my garage. I wanted to make sure the the line was dead before I started removing it.

I used pic related hooked up to the voltage and common leads to the 120v input line to check for voltage and I ended up tripping the breaker and melting the leads

I thought volt meters are supposed to have a very high resistance.

>>968855
you probably touched your two leads together, shorting the circuit.

>>968855
upon further inspection the device says "CAT II 600v, 200mA max". Is this OK for 120v wall outlets?

Is there a way to calculate the resistance of the device?

>>968858
couldn't have done that, it's an old lever type breaker box with the two input leads on opposite sides.

>>968855
if you put the leads in com and 10adc there is a only a small shunt resistor in circuit which will allow massive current to flow.
you should have had the leads in com and v/ohm/ma socket. leave them in this position when not in use to avoid accidents.

or its just a cheap shit meter

>>968860
I can almost guarantee he did this. >>968855
Use the volts side of the meter, not the amps side. Current is meant to be read in series with the load, volts parallel with that meter. If you read current in parallel you have created a dead short, and at best only blew a fuse in the meter

>>968876
>>968855
Also, with these cheapy meters, don't turn the dial while measuring, and don't test resistance on a powered circuit.

>>968818
Hmmm righto, transformer is a TM50-2 do you think 0.3A difference in current would be enough to saturate the transformer? Also there is a 0.5A fuse on the AC side.

Also the goal of the project was to get a stable linear supply, though going to switching would be advantageous control and efficiency wise I think I'll stay linear for this run.

>>968840

it's impossible to tell from your child-like drawing where the splice will be located. if it's outside a box, then you gotta make access to the copper impossible, so use 4 pieces of heat-shrink, one per wire, then a thick one over the other three. heat-shrink should be longer than the splice by about 3/4-inch in both directions.
if it's inside a box, a whole bunch of options are available: crimps, wire nuts, insulated spade terminals, screw-terminals, solder terminals strips, etc.

>>968893
> Hmmm righto, transformer is a TM50-2 do you think 0.3A difference in current would be enough to saturate the transformer?
I can't find any specs on the transformer, which makes it quite hard to even guess.

If the core reluctance is particularly low, even a small DC current can cause flux to increase continually until saturation occurs, at which point the core stops having an effect upon the inductance and the current is limited only by the resistance of the winding.

Toroidal transformers are particularly prone to this, as the core is perfectly continuous. Split cores and laminated cores have some gap which counteracts the effect.

> Also there is a 0.5A fuse on the AC side.
Well, at worst you'll only blow the fuse rather than the transformer. The main thing is not to assume that you can use the rated capacity of the transformer with an unbalanced load.

Try hooking the transformer to a test circuit (two rectifiers and some power resistors) and measuring the current on the mains side.

>>968916
>A more fundamental issue is that if the current drawn from the +ve and -ve outputs isn't balanced

this doesnt make any sense to me. it's a 2-wire secondary, using a voltage doubler to generate + and - outputs. the total current out of the transformer will always be the sum of the output currents. there can be no unbalanced anything.

>>968902
The connections will all be on the inside of the box. Now could the joints be spliced and soldered and heat shrinked? I never hear much of power cables being joined in that way, but it would save me money. And if that is too unsafe the crimp connector sounds like the neatest and simplest option (other than soldering)

>>968926

heat shrink is fine for power cables. if you get brand-name heat-shrink, it's rated for 300V.

Sup guys, so i'm making this project that consists in a light that is turned on and off by the sensor, so i got this scheme as a timer, but for some reason it doesn't work, what is wrong with it?

https://gyazo.com/cc1287deec7ed0de073decafaaf866b2

Thanks

>>968975
So is your sensor SW1?

>>968990
Yes

Thanks

>>968975

not gonna try to diagnose your circuit thru the internet, coz that's even more hopeless than doing it by phone, but i can tell you there's a stupid newbie mistake on your circuit. there's a pot between pin 7 and the +12V. if that pot is ever adjusted to zero ohms, then it can cause a transistor inside the 555 to die, and then the 555 wont work. so, your 555 may already be dead. try replacing it, but add a 1K resistor between the pot and 12V before you do it.

>>969142
I wanted to simulate this circuit
https://gyazo.com/5a64a8c34b545532bc08d2f0b618c895

I thought that i had that right, but it looks like not, is it a problem of the circuit that i'm trying to copy or is it my problem?

Thanks

File: halfwave.png (15KB, 948x476px) Image search: [Google]

15KB, 948x476px

>>968921
Here, "unbalanced" refers to the situation where the current drawn on one half-cycle is consistently higher than on the other, meaning that the current waveform has a DC component (i.e. the mean current over a complete cycle isn't zero).

>>969158

the circuit itself is correct. when the sensor puts a voltage on the 100K resistor, it turns on the transistor, which creates a negative pulse going into pin 2 which starts the monostable period, turning on the relay for a certain period based on the resistance of the pot.

but if you try to simulate it, the simulation may not work properly even if the circuit is correct. that's probably because the simulator expects a clean digital pulse on pin 2 of the 555 but it's gonna get a funny spike instead.

anyway, you can wire it up on a breadboard just as quickly as using a simulator. i would strongly suggest adding a 1K resistor in series with the pot so you cant short pin 7 to V+. also, the 100K on the sensor input is probably too high; i'd change it to 22K to make it easier to turn on the BC548.

>>969229
Hey, thank you for helping, that helped a lot, thank's a lot.

One more thing, i want to connect this to a 12v led strip, what do i have to change? Because the original circuit is built for a 220V lamp.

Thank you

>>969240
> i want to connect this to a 12v led strip, what do i have to change?
The simplest change is to replace the 220V lamp and the 220V plug with the LEDs and a PSU. I.e. just the stuff on the switch side of the relay; everything else stays the same.

The alternative is to change Vcc from 5V to 12V, ditch the relay (and flyback diode) and put the LED strip in its place. Depending upon the current drawn by the strip, Q2 may need to be upgraded. This has the advantage of getting rid of the relay, which may account for a significant portion of the circuit's cost and/or size.

File: now with LED strips.gif (26KB, 725x726px) Image search: [Google]

26KB, 725x726px

>>969240

if you use the 12V version, you can just replace the relay with the LEDs. you'll need a stronger transistor capable of putting out the 200mA or whatever the strip needs. a TIP30, for example, can handle 5A.

for the 5V version, you can replace the lamp by a xformer and LEDs.

>>969268

fag, you stole my answer.

I have an idea for a very long-lasting 'candle' type thingie for power outages and the sort.

Anyways, I have a dozen of these LEDs and I have no idea what their wattage, voltage or anything is. The diode function on my digital multimeter will dimly light up the LED but it doesn't give me a reading, same with all the olmmeter readings. It lights up when hooked up directly to a 6v lantern battery.

How can determine the specifications of a random LED? I don't want to burn the LED out by supplying too much power to it and it already produces an alarming amount of heat when hooked up to the 6v lantern battery.

They are about 5mm across, blue-white type led.

>>969306
>I have an idea for a very long-lasting 'candle' type thingie for power outages and the sort.
If it's not more efficient than solar panel > battery > LED bank it's pointless.

>>969313
This is for personal use, I'm not going to try and market this. But I don't want to have a shitty product for home use either. Solar panels are expensive, carbon-zinc battery piles, used lead-acid 12vt cost next to nothing.

Also I'm just a novice of electronics, even though I'm a competent electrician. I want to design an efficient product.

>>969325
you can order a ~200mA solar panel from china for $6 shipping inclusive

>>969326
20 amp-hour battery pile is like $2 from down the street.

I like photovoltics but this isn't an application where I care for their use. It's an electric candle. I don't want expensive rechargeable battery chemistry. I don't want any sunlight in my basement for it to work.

It's gonna be a big, fat, cheap battery hooked up to an LED and it should last a long time.

I'm not trying to fucking sell you this thing, I just want to know how I can figure out the specifications of these nice LEDs I hve for the purpose I want to use them form.

>>969306
> It lights up when hooked up directly to a 6v lantern battery.
What's the measured voltage across it in that state? If it's more than 4.2V, it may have a built-in current regulator.

> How can determine the specifications of a random LED?
Find the part number. Otherwise, the only specifications you're going to get are those which you can measure, i.e. the forward voltage and its emission characteristics. The only way to estimate the maximum current is to find the current at which it fails.

>>969325
The simplest approach to 12V LED lighting is to just buy standard bulbs and fittings (mostly MR11, MR16, or G4 sockets, sometimes E27). MR11/MR16 are normally spot lights (narrow beam angle), while G4 and E27 have more variation.

Most of those sockets support different voltages (MR11 is 6V or 12V, MR16 is 12V or 24V, most E27 bulbs are 120V or 240V but 12V bulbs exist).

However, such bulbs are designed primarily for 12V RMS AC (i.e. being fed from a 12V transformer) so some bulbs may not work with 12V DC.

Caravan/RV lighting is another off-the-shelf option.

>>969329
> I just want to know how I can figure out the specifications of these nice LEDs I hve
Unless there's a part number, you can't. Visual identification won't work; it's common to use a single package type for many different LED types.

Just throw them away and buy something for which you have the data sheet. It's not like they're expensive.

>>969354
Measuring across on the 6v lantern, it's 5.6v but it also measures 4.5 amps, which is bonkers for a little LED light. I think my multimeter is just shorting it out.

I'm not going to buy the idea that 'Unless the box says it's this or that you'll never know!' This isn't true for any resistor, capacitor, diode or fucking transistor on the planet, I don't know why it would be different for these LEDs.

>>969418
> Measuring across on the 6v lantern, it's 5.6v
Measuring across the the LED while the battery is connected? If that's the case, it would have to include some form of current limit. The maximum Vf for any LED is ~4.2V (those are UV; white LEDs are typically UV LEDs with a phosphor), and current typically doubles for every extra 20mV, so there's no way a bare LED will survive 5.6V for even a few seconds.

It may have an active constant-current circuit, or a series resistor chosen for either 5V or 6V operation. If it's a resistor chosen for 5V, then 6V will roughly double the current (you'll be dropping 1.8V across a resistor chosen for 0.8V). Which is survivable for short periods, but don't expect them to last long.

> but it also measures 4.5 amps, which is bonkers for a little LED light. I think my multimeter is just shorting it out.
Measuring the current between the battery and the LED while the LED is lit?

5.6V x 4.5A = 25W.

A 5W LED bulb produces roughly the same light output as a 40W filament bulb, and those use multiple dies.

You aren't just measuring the battery's short-circuit current, right?

>>969418
> I'm not going to buy the idea that 'Unless the box says it's this or that you'll never know!' This isn't true for any resistor, capacitor, diode or fucking transistor on the planet
It's true for all of those.

The only way you can determine the maximum current or power of a component experimentally is to test sufficient numbers of them to destruction. You need to test at multiple values of current (or power) and have enough data points at each value to be statistically-significant.

>>969268
>>969272
Thank you so much for helping.

I liked the ideia off getting rid of the relay, but what would turn on and off the led strip after the time i wanted it to? Is there any downside of replacing the relay?

Thanks

>>969505
Forgot to mention.

If i use the 1st opton, taking off the relay and upgrading Q2, i only need 1 Power Supply/Transformer? With the relay i need 2 so i can give the 5v to the circuit and convert the 220VAC to 12v dc to the led strip right?

>>969505
> but what would turn on and off the led strip after the time i wanted it to?
The timer. The original circuit turns the relay on for a certain amount of time, and turning on the relay turns on the LED strip.

The relay is there because it provides a simple and safe way of controlling mains voltage via a logic circuit. A triac or MOSFET would also work, but doesn't provide isolation, and any mistakes in construction create a safety risk.

> Is there any downside of replacing the relay?
Flexibility. With a relay, you can control any AC or DC load up to the relay's current and voltage limits. Also, the relay provides isolation.

>>969506
> If i use the 1st opton, taking off the relay and upgrading Q2, i only need 1 Power Supply/Transformer?
Yes. The 555 can be powered by anything from 4.5V to 16V, so you can power the timer circuit from the same supply as the LED strip.

> With the relay i need 2 so i can give the 5v to the circuit and convert the 220VAC to 12v dc to the led strip right?
With the relay, you can either switch the 220V AC input to the PSU or switch the 12V DC output. With the latter, you could use the 12V DC to also power the timer circuit. But there wouldn't be any point having a relay in that case.

File: leddrive.png (1KB, 234x183px) Image search: [Google]

1KB, 234x183px

>>969506
Something that needs to be pointed out: many of the replies are based upon the assumption that the "12V" LED supply is a just a 12V voltage source and the LED strip has its own current regulation.

If the supply is a constant-current source and the strip is just a chain of LEDs (i.e. current regulation is provided by the supply rather than the strip), you can't power the timer with it, or just put the LED strip where the chain was.

You can probably still skip the relay and control the strip directly with a transistor. The main thing is that the transistor needs to be capable of handling the current and voltage and have enough current gain.

>>969690
Thanks again.

So the new Q2 transistor will be a TIP30 right? Since i want to take off thhe relay.

The transformer, i would plug it into the timer circuit terminals right? The ones on the diagram?

Thanks

>>969694
The strip is one of those that you see out there, like this one: http://www.ebay.com/itm/5m-Super-Bright-12V-2A-3528-SMD-Led-Flexible-Strip-Home-Decoration-Light-EU-Plug-/331790376759?var=&hash=item4d403f5f37:m:mguzWUeK8IO_EzMj9vcyuvQ

I wanted to connect the power supply that comes with the led strip to the timer circuit, and then connect the strip to the timer circuit.

Thank's so much to both of you :)

>>969704
The ebay listing gives absolutely no relevant technical details. The safest option is to keep the relay.

If the strip is relying upon the supply to perform current limiting, the supply won't be able to drive timer circuit, and the timer circuit won't be able to drive the strip.

The only thing we actually know is that the supply+strip combination can be powered from the mains. The original circuit (with the relay) achieves that.

>>969767
But i will need a 2nd power supply with the relay :/

The cable is just like this, with 4 cables at the end, http://www.builderlights.com/content/172237/BL.JPG

I just found this tutorial, what do you guys say? http://www.instructables.com/id/Hack-your-LED-Strip-with-a-Motion-Sensor-and-Timer/

>>969456
You might be correct, but when I thought about what I'm trying to do, I really just need the 'minimum' for the LED, because it's meant for longevity, so I'm just going to add D-cells in a series until the LED is suitably bright.

>>969447
Thanks, this is actually really informative, but on trying out the LED, it's not even very bright on the 6v battery, it doesn't even get hot. The LED is in clear plastic, I don't see a resistor and judging by the heat output I don't think it's resisting anything.. I think I'm shorting it with the multimeter cause 25 watts through an LED would probably flash-blind me and it's pleasant to look at, I need to stick this thing into a breadboard before I fuck around with it anymore.

>>969859

> another kid doesn't understand LEDs

File: IMG_0945.jpg (642KB, 1887x2384px) Image search: [Google]

642KB, 1887x2384px

Doing my first AC project...think i made a mistake

I'm making a big project box that houses a 6 socket AC box that feeds off of one cable. That cable will be plugged into a wall socket attached to the house.

House power will be connected to leftmost prong in my pick, then the other prongs each connect to the hot side of an individual outlet (I have fuses for safety and relays for control via arduino).

My question is...is this configuration (series) okay if its ac? or should I wire each outlet parallel to the house supply

Hoping to not die