/ohm/ - electronics general

last >>1166999

>>1171982
>dismissive
Stop being retarded and people will stop calling you retarded.
>>1171983
I already posted the schematic and power requirements you illiterate retards:
>>1171160
>>1171390

>>1172022
>>1171982
I apologize to you, I was referring to the other guy. If people weren't so retarded they would come up with something that could do what I need to do, but they can't, not even a little bit. Funny how that works.

>>1172025
>>1172022
as others have said there's better ways to do what you seem to be doing. but you are unwilling to do that and just wanna keep on doing that dumb shit you're doing
kys my dude

File: IMG_0406.jpg (1MB, 2448x2448px) Image search: [Google]

1MB, 2448x2448px

Hi /ohm/ could anyone help me identify the damaged part in this headphone amplifier? It's identicle to the adjacent non-damaged part and has the markings 1074 52704. I was hoping to find the part and repair it. Please forgive me if this is a dumb question.

>>1172044
replace E16

>>1172044
It's a tantalum capacitor, I can't read the value, try finding a circuit schematic or maybe someone with the same model who can take a picture for you.

>>1171979
B

>>1172072
The "107" part of the name suggests 100µF, but there are a lot of other numbers there too. "527" could mean 520µF, but this isn't a standard value so I doubt it. If you can't find the value of the existing caps you can remove an existing one and test it through a variety of ways you can learn about online, but it's much easier if you have an oscilloscope, LCR meter or capacitance multimeter, or Arduino.

>>1172077
>or Arduino
Hook it up through an analog port and print the value?

>>1172096
If by that you mean put a voltage across it in series with a resistor and read the voltage across the capacitor via the built-in ADC and use this to calculate the capacitor's time constant, then yes.

Where can I source cheap pots?

>>1172125
ebay / china

Has anyone used a recovery diode as an active circuit elenent? Still seeing if I could "extend" the breakdown region of a mosfet and didn't account for the body diode in my voltage orientations. It may permit me to do silly shit with orienting and driving the mosfet. Itd be the same as a diode in parallel so I think that might solve two problems I have at once

>>1172109
If it's 100uF or anything near to it, you could measure the time constant with sufficient accuracy by a normal multimeter and a reasonably small resistor.

Does it make sense to buy windshield wiper motors for electronics projects? As in, do they provide a better power/price ratio than just buying motors on electronics e-shops?

When would I use a PIC controller over an arduino?

>>1172297
Roughly in the same situation where you would use an AVR over Arduino.

>>1172314
meaning it's the same thing?

>>1172297

PICs come in a very wide variety, so for example, if you wanna built something that'll last a month on a coin cell, you'd pick a PIC coz an arduino would drain it in 2 minutes. or if you want something really tiny, that'll fit inside a bottlecap, once again you'd pick a PIC.

kind of a dumb question, but Im too broke to afford a pricey multimeter. Is there anything like a 'PC multimeter' that plugs into the usb port and gives you a readout in the screen?

I guess I'm looking for cheaper but reliable alternatives.

>>1172372
How about one of these fuckers?
http://www.ebay.com/itm/252811751877?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

File: parts.jpg (256KB, 1600x1280px) Image search: [Google]

256KB, 1600x1280px

newfag to electronics here. just got into it, reading up and watching vids. Are there any kits out there that can get me a good number of resistors, LEDs, capacitors, switches, ect for a price under $50? preferably on ebay or amazon. pic related

>>1172413
>ebay
>"learning electronics kit"

File: card7a_rev1.png (373KB, 632x692px) Image search: [Google]

373KB, 632x692px

I bought a teensy
Is there a getting started guide or anything
The website is total trash and all there is is a pinout

>>1172420
Teensy 3.2?

If you like using the arduino IDE there's teensyduino on the pjrc website. Other than that some youtube videos on it.

>>1172413
BUY BULK

File: s-l1600.jpg (108KB, 1200x1200px) Image search: [Google]

108KB, 1200x1200px

>>1172372
>too broke to afford a pricey multimeter

http://www.ebay.com/itm/A-NEW-CEN-TECH-DIGITAL-MULTI-METER-CEN-TECH-7-FUNCTION-DIGITAL-MULTI-METER-/262966034153

$4.90 - Free Shipping

>>1172438
>literally harbor freight trash
>soldered in fuse
>no auto off

>>1172440
Sure, but one of the product pictures is the statue of liberty in front of an American flag.
USA FUCK YEAH! BUYING A DOZEN RIGHT NOW! MAGA!

>>1172438
I have one like that. It's my first ever DMM and its still ticking. I also have a uni-t 39 that has some intermittent issue where it doesn't measure shit unless i push the dial down hard (something to do with contacts i guess). So now i have the eevblog one and its great

>>1172440
>>soldered in fuse
>>no auto off
>blowing fuses by improper use
>too stupid to turn meter off
> < $5.00

>>1172438
looks like a good way to start a fire

File: Screen Shot 2017-05-05 at 6.52.16 PM.png (172KB, 1572x738px) Image search: [Google]

172KB, 1572x738px

>>1172204
Small enough to ignore any leakage current? That would be nice, but surely you could just run any smaller cap across a voltage source and measure the current to figure out the leakage resistance, and then use this with the large-resistor RC charge/discharge test to figure out the capacitance. That is at least if the series resistance is negligible in comparison to your load resistance, which it might not be if your load is small enough.

Technically, if leakage resistance is just due to constant resistivity of the dielectric, each dielectric material would give the same time constant regardless of the capacitance of its capacitor or the voltage, and this can be calculated solely from the dielectric's permittivity and resistivity. This, along with a known capacitance, can give you the leakage resistance of a capacitor, theoretically speaking of course. Inversely, you'd be able to calculate capacitance just from the known time constant and measured series DC current. Too bad getting concrete values for resistivity and dielectric permittivity is nigh on impossible, so if you wanted to make a capacitance meter this way you'd need a lot of measurements on different cap types to figure it out. Have my spreadsheet.

>>1172438
> less than $5
Have you tried Aliexpress lately? Clamp meters for $6.50, trash DMMs like that for $2.70, and an analogue multimeter for $2.40. While I don't think the 9V can burn your house down easily, but no way in kekistan am I ever sticking those into a wall outlet.

>>1172539
>kekistan
opinion descartes

>>1172541
I'm a Socrates fan myself.

>>1172539
>>1172541
Kekistan, population: (You).

>>1172044
They MAKE capacitor, inductor, etc testers.

http://www.ebay.com/itm/Transistor-Inductor-Capacitor-ESR-Meter-MG328-Digital-LCD-Tester-Case-/311765370403

Mine's like that. It's usb powered, and needs a battery in the holder, but doesn't actually run off battery. Dunno. The inductance usually only picks up at 10-30 uH, but capacitance goes down to tens of picofarads. It actually can do PWM stuff and other weird functions if you wanted it to, by holding down the trigger button.

So I found my old game gear and when I turn it on the screen doesn't work but you can tell it's trying to. There's a little backlight and like a green line at the bottom. How easy is it for a somewhat resourceful person to change capacitors with no experience.

>>1172584
If you have the basic ~$40 of soldering equipment, a general idea for what not to do to blow up electronic components, and a low score on the recklessness gauge, you should be good to go. Let's just hope that its the through-hole caps that went and not the SMDs. Pics?

>>1172584

the 'replace the caps' meme is fairly new. old caps on old machines that dont use switching power supplies tend to last for a very long time, so dont assume that changing caps will work.

but if you wanna have a go, try this instead: wire new caps in parallel with the old ones temporarily. since the problem with caps is that their capacity drops with age, this should fix it, if that was, in fact, the problem.

hi, I have a quick question, simple and all but I have been thinking about it for a while.

I whant to heat something using DC. I have got two power resistor. 10ohm 5W and 6.5ohm 5w.

What is the advantage of putting in serie or in parallel?

>>1171979
I'm not much of a tinkerer and my background is in math, not any engineering discipline. Now that I'm out of Uni and don't have access to their tools anymore, what's are some cost-effective strategies of getting my hands on equipment I need to use for only a couple of minutes? I live in Chicago, so I probably have options, but I don't even know what to look for. I'm currently just renting an occasional tool from home depot if it's out of my price range.

>>1172722
Auto-parts stores loan some tools.
They just require a deposit that you get back.

As far as homowner type tools, if you're not using them much, get a mal-wart or pawn shop cheapy.

Had a TV repair shop let me use their oscilloscope once.

File: Ohms law.gif (8KB, 405x369px) Image search: [Google]

8KB, 405x369px

>>1172690
Series would allow you to use a higher voltage.
Parallel would allow you to get more power from a lower voltage.

What are you powering them with?

>>1172730
okay cool. I'm not really familiar with etiquette in this area. There is actually a large number of various workshops within a block of me. When they aren't busy should I just walk over and introduce myself?

>>1172690

series or parallel is irrelevant, as long as the supply voltage matches.

in series, you'd get 16.5 ohms. to get 10W, you'd need 12.8V, at a current of 0.775A. but because the resistors are different, you'd get an imbalance, with 6W across the 10-ohm and 4W across the 6.5-ohm. so you'd have to drop the voltage below 12.8 to keep it from burning.

in parallel, the resistance is 3.9ohms, so to get 10W you need 6.2V at 1.60A. now the 10-ohm gets 3.8W and the 6.5-ohm 5.9W. so, this time it's the smaller resistance that's in trouble and requires you to drop the voltage, and so in neither case can you get the full 10W from two 5W resistors coz they're not equal.

another example of why race mixing is bad.

File: DSCN5123.jpg (2MB, 2000x1476px) Image search: [Google]

2MB, 2000x1476px

Electric kettle. Any reason why this would happen after months of use? The only thing I can think of is the resistance of the heating element decreased, but why would that happen? Is it safe to resolder?

>>1172731
I think i will take 5v. Thank for the answer.

File: 20151226_182145.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1171979
I've got to install a new mast for my house I bought. I'm allowed to do all the electrical and it needs to be up to current MN code (follows the NEC). I have the most recent copy of the NEC but can't find a whole lot of specifics on mast installation.
Has to be tall enough so snow doesn't accumulate, is there a recommendation for height? It needs to be secure, what is the best way to do this for a plastic tube sticking through your roof? Guy wires?
Does the wire need to be continuous from the meter to the service point? If I can, what splicing is acceptable?
What Guage do I need for standard one phase residential?
If there is not enough wire from the line, do they have to re-string a whole new one, or just splice an extra length on?
I'm not a retard, just trying to do it right but can't seem to find explicit information easily, the NEC is a mess imo

>>1172746
The wire was probably crimped badly and failed after some use.

>>1172044
>markings 1074 52704
Bit difficult to say, but I'd guess it's actually 107A.
They look like B-size 100uF tantalum capacitors made by AVX, probably rated for 10V.

>>1172722
Make friends in computer/TV repair shops. One of my old classmates works at a place that sells and repairs computers, their workshop has all kinds of electronics tools and he lets me use pretty much anything as long as his own boss isn't around, which hasn't happened yet.
>>1172733
Be friendly and if they look busy, wait for a calmer day. If they find you annoying, they're not going to let you borrow anything.

recommended bench PSU and oscilloscope for beginners?

>>1172746
>Is it safe to resolder?

No. The melted part is molten wire - not solder.
It wasn't soldered from the factory, it was crimped.
The heat involved with the element would likely melt any regular solder you were able to apply.

The other terminal looks a bit on the edge of failing too.

It could be poor contact with the flat blade or just bad design.

I'd probably toss it since they are both bad but if you want to try to repair it:

Get new good quality terminals
Cut the ends of the wires
Carefully strip the insulation back a bit
(this high-temp insulation requires very sharp tools)
Crimp the new terminals
Attach in place of old terminals

If it fails again, throw it out and get a different brand.

>>1172754
Makes sense.
>>1172793
>The heat involved with the element would likely melt any regular solder you were able to apply.
Where do you draw this conclusion from given that 1.) the plastic in and around the terminals isn't melted, and 2.) the ends on the element side are indeed soldered?

>>1172796
>the ends on the element side are indeed soldered?

they're welded so they can withstand 1000 degrees.

>>1172790
Do your own PSU, it's fairly easy.

Any UK fags in here?

I'm fairly new to this sorta shit and was gonna get a teensy LC to try and build a little control panel thing since it seems like basic functionality is super easy and still gives me room to experiment and learn but am not sure where to buy shit from.

I just got my little breadboard kit and soldering iron and shit off amazon but have no idea where is the best place to order swtiches, buttons, pots, etc. from.

Question: Do I have to use a laminator when using photoresist film? Can't I use my fingers to thoroughly adhere the film?

https://www.aliexpress.com/item/Professional-30cm-Photosensitive-Dry-Film-Replace-Thermal-Transfer-PCB-Board-Longth-1M-Brand-New-no-wrinkle/32508901573.html

I want to step up my game but I want to avoid buying a laminator if I can.

>>1172790
Rigol DSZ1054Z is the way to go for a scope, it can be hacked up to 100MHz. I have one and I'm in love with it. No Idea about the PSU though, I've been using a laptop charger and LM317 based PSU for years. I'm in the process of making a new PSU with an audio amplifier transformer which I plan to rewire for multiple taps to improve the efficiency since I plan on using linear regulators to avoid noise. I would suggest this, specially if you're a beginner >>1172818

>>1172821
Why?
You can easily get a used laminator for 20 bucks.
Also comes in handy if you want to experiment with toner transfer.

Might need some modification to take thicker pcbs though.

Hey is this a dumb idea


Car inverter keeps blowing fuses from drawing too much through a 15a fuse through the 12v socket.

Theres another 12v socket next to it. Could I plug another lead into this one, connect both leads to their respective studs on the inverter, and not blow any more fuses?


Purely a temporary fix until i get the supplies to hardwire it like a normal human bean

>>1172796
>Where do you draw this conclusion from

I just pulled it from my ass.

Solder it back and everything should be OK.

Sorry for giving advice when I had no idea of what I was talking about.

>>1172850
I already use the toner transfer method without a laminator using the electric (vitroceramic) stove in my kitchen.
The thing is I can't buy it used, so I'd rather not spend the money if it isn't crucial.

>>1172861
Are you sure botb sockets son't work on the same fuse? But yeah, you could probably parallel them to share the load.

i feel like a fucking moron. i figured you could cut a 3000w heating element in two and you get 1500w. i had this fleeting thought in the back of my head that something wasn't quite right about this idea but i bought a 3000w heating element anyway. oh well, 4.15€ isn't too expensive for a lesson.

>>1172868
Yea i already pulled the blown fuse while i used the good one with reduced load. Theres actually 5 of these sockets all on their own line.

Just wanted to make sure it wouldn't overload the inverter or something silly.

>>1172869
buy another one and fuse them all three in series

>>1172872
i just ordered the 3000w but learned about my mistake too late to cancel the order
it's cheaper to buy lower power ones anyway

>>1172878
yes. If you order a new 3000W heating element then you can solder it in series with the two halves of the first one you bought.

how can i get a good gauge on how bright a led is before i buy it? i understand lumens and candelas mathematically but they're meaningless to me. i'm mostly concerned because i'm setting up 5 rgb leds as a multicolor bar display and i want to match the r, g, b brightnesses so one color doesnt swamp the others out.

led in question is clx6c-fkb

>>1172882
3 1000w or 2 1500w are cheaper than another 3000w one on aliexpress

>>1172883
>>1172885
also 2 3000w elements would be too long for what i want to do

>>1172821
>I want to avoid buying a laminator
Get spray or paint resist then. They're shit, though. Toner transfer method is another option.

I'm trying to turn my old computer PSU into "bench" PSU.

Is the cooling fan controlled by something on motherboard(which would be logical) or is there a circuit inside of PSU that does it instead? If it's controlled by mobo then where is it getting information about current temperature of the PSU?

>>1172820
Farnell

>>1173026
ATX PSUs dont' have any external fan control.

Usually no internal control either - it will just run at constant speed.
You can check inside the PSU and look at the fan. If it's just 2 pins, it's very unlikely that there's any fancy speed control.

>>1172983
I already use the toner transfer method, it works but making sure the toner sticks properly and covering small holes with permanent marker is a bit of a hassle.
I'll probably get a laminator because all the people I've seen use one.

>>1173039
It has 4 pins. But yeah, I guess I can make it run at 100% speed whole the time.

>>1173044
No need to. If its a PWM fan it has the monitoring hardware required to regulate it too. Just leave it as it is.

>>1173053
I've looked it up.

There are only two pins on the PSU and while it's physically impossible to reach them with 4-pin connector, I think that's the on-board cooling solution for the thing(also known as "connect -12V line to the fan").

I think I can make a controller for this, the problem is that it uses some unmarked output transistors which I assume are IGBT and even though I've found a schematic of that PSU(can put those pics in if you want), they're also unmarked there, so I can't tell what temperatures are they supposed to work in, and in turn, I can't figure out what temperatures it should work in. I mean I can get the temperature by gluing thermistor to the radiator and assuming the transistor is few degrees hotter, but I can't tell what is the upper limit for temperatures I can allow them to achieve.

Any ideas on what range of temperatures are we talking about here?

>>1173128
Why bother? Let the fan spin.

File: diagram.jpg (105KB, 1488x814px) Image search: [Google]

105KB, 1488x814px

I'm doing a project this summer that requires the use of 8 stepper motors. The stepper motors that I want to use have 4 steps on their stator. As such each stepper motor needs 4 wires to control the step sequence. This is no big deal, I've done this using 4 outputs from a micro controller.

My issue is that I need to control 8 of them with only 4 outputs from a micro controller. A diagram of what I want is attached. Does anyone know what component could accomplish the behavior of the "MUX" in the picture?

For example, If I send the 8-bit signal "0010 0001" to the first "MUX", then its going to take the 10V DC supplied to that "MUX" and pass it through to the first and sixth stepper motor.

>>1173188
Transistors. If you have one additional pin available, you can simply swap between them with that, ie:
pin 5 low -> Control first four motors
pin 5 high -> Control second set of motors
Otherwise you'll need to split them into three groups and use two sets:
pin 4 low, 5 low -> Control first three motors
pin 4 low, 5 high -> Control second three motors
pin 4 high, 5 high -> Control third three motors (one extra output, can be ignored)
pin 4 high, 5 low -> Three extra outputs.

>>1173199
But the thing is, none of the motors will be acting in unison, so I need to be able to individually control them with each clock cycle of the processor

File: file.png (53KB, 563x558px) Image search: [Google]

53KB, 563x558px

>>1173188
this will work. hopefully you're not too new to interpret it into a proper circuit because i'm too lazy to draw anything better.

assumes you have unipolar steppers.

>>1173188
32 bit shift register SIPO with latch and tri-state output. Make from four 74595 or equilavent
>http://assets.nexperia.com/documents/data-sheet/74HC_HCT595.pdf

pin 1) data
pin 2) shift clock
pin 3) latch clock
pin 4) output enable (may not need if a coil is always on)

anyone ever purchased anything from banggood or should i just stick to ebay?

Trying to figure out why Multisim would have GHz ringing and not be able to suppress it. I have suppress numerical oscillations, I can change very damn setting, but it still detects near instantaneous ringing, and can't do shit about it. I can turn the error tolerances through the roof and it still won't go away.

>>1173335

Because it is modelling a circuit with unrealistic components which would cause GHz ringing if they could really exist?

How much is too much fuse and how do you calculate it?

I have a 1000 watt inverter for my truck, so its 12v. Doing the maths it should draw 125 amps. I was going to use a 130 anl fuse. Calling the local hoodlum supply for stereo amps they recommend a 200 amp fuse.

I honestly know minimal electrical so I'm not sure how sound that advice is, or why its acceptable to go over the amps. Obviously a little is necessary with standardized fuses.

>>1173398
A fuse is there to protect against a catastrophic event.
130A is less than 5% over the calculated 125A, and wouldn't call for the fuse to blow. It may be even higher at initial power-on.

I'm not familiar with high current fuses but on regular componants, you only want to use 50%-80% (depending on who you ask and how reliable it needs to be) of it's rated value. Using that, a 200Amp fuse for a 125Amp circuit sounds reasonable.

>>1173398

what size wire are you using?
the fuse is to protect the wire.
if the fuse blows the inverter is already gone.
(that's why the fuse blew)

>>1173404
Alright i see. Seems kinda fishy to just guestimate like that. Can equipment generally take more amps than normal? Is heat the only real side effect?
>>1173405
I was going to use 0g, or slightly smaller in case i need a bigger inverter in time.

Im not sure what you mean by the inverter being gone if it blows. Everything after the fuse should be saved no? I know houses use breakers to prevent the wires getting hot from too much wattage being drawn which also applies here, but drawing too much isnt a big factor here. Im not plugging in many power hungry devices

>>1173412
the fuse doesn't protect the inverter, it just breaks the circuit if the inverter fails short. the inverter circuitry itself may be protected by temperature and current sensing controls (or not).

File: tmp_9331-20170507_112524628080266.jpg (863KB, 2048x1536px) Image search: [Google]

863KB, 2048x1536px

I'm designing a UV exposure box and I have some questions.

1) Is a 24 watt 5m UV LED strip enough for this purpose? It's a double sided exposure box, so it will have 12 W on each side over approximately 3/4 foot^2 (25cm^2) square of area.

2)LEDs have a very wide emission angle, does this cause problems when exposing the boards?

3)What is the ideal distance between the LEDs and the glass sheet on which the board is sitting?

Pic related, my design.

Can anyone recommend a micro controller that can handle 6 encoder driven pid loops at once?

I have magnetic quadrature encoders on my motors that report 4096 steps per rotation. The motors all run around 130 rpm (the encoders are mounted after the gearboxes). so ideally I would want to be able to collect something like 50000 pulses per second on 12 pins with a good level of accuracy.

Ideally I'd use lower pulse/rotation encoders but it's not an option for this project.

Most of my fiddling has been with arduinos and raspberry pis so I'm not sure what to use in this case. accurate pwm generation would be nice as well.

Thanks for any responses.

>>1173478
Alternatively if anyone has any resources on what sort of processor speed or loop time is required to measure an encoder giving x pulses per second with a certain amount of accuracy that would be tremendously useful.

So I got a bunch of ATtiny13A chips from aliexpress and I can't seem to program them.
avrdude says target doesn't answer.
I figured maybe they are set to take an external clock so I programmed an arduino to generate an 8MHz clock signal and fed that to the clock pin. The clock signal is 5Vpp.
I have my usbasp programmer connected to it as well, having it set so the ATtiny is powered from the programmer, not the arduino. Just the grounds are connected.

Now the strange thing is, when I have the usbasp connected only to the attiny and not plugged in the computer, and I feed (only) the clock signal from the arduino to the clock pin of the attiny (pin 2), then VCC on the attiny (pin 8) starts putting out 4V, powering the usbasp.

Is it a fake attiny perhaps?

>>1173458
Simply put, you'll want to maximise the intensity of the UV on the boards. The simplest way of doing this is by covering the inside of the box with something that reflects UV light, unfinished metal should do. This means that regardless of the angle of the LEDs, the UV light will bounce about until it hits the board. This is ignoring losses through the small absorptions and possible photoelectric effects, but it shouldn't be too much of a problem. I hope you know how UV light boxes work and what kind of intensity and wavelength they require, because I've got no clue. But I'm pretty sure that intensity will only change the time it takes to expose the boards.

>>1173484
I got some cheap 13As from Ali as well.
I programmed them without any external clock just fine, using just a cheap usbasp clone.

Have you checked the fuse bits?

Anybody know if it's viable to electroplate extra copper onto your traces after etching them to get them to handle more current, or will the layer not be consistently thick or solid?

>>1173494
Depends on your idea of viable, but if you do it manually one track at a time, I fail to see how this is going to be a better idea than a) solder-coating traces (if you need only relatively small improvement), b) soldering copper wire directly to the traces or c) using thicker copper to begin with.

>>1173493
kinda hard to check fuse bits when all avrdude says is
>avrdude.exe: set SCK frequency to 93750 Hz
>avrdude.exe: error: program enable: target doesn't answer. 1
>avrdude.exe: initialization failed, rc=-1
in response to
>avrdude.exe -c usbasp -p t13 -U lfuse:r:low_fuse_val.hex:h -U hfuse:r:high_fuse_val.hex:h -v -B10

Also I'm a complete noob at this so I'm not even sure I am trying to read the fuse bits correctly. Just trying to google my way through this.

Got a cheap usbasp clone which didn't let me program the tiny13, so I upgraded the firmware using an Arduino as ISP, after which it did at least let me set the clock frequency (which didn't work before), but still the tiny13 won't respond to the programmer. With or without the slow clock jumper.
While I had the arduino configured as ISP I tried programming the tiny13 with that but that also didn't work.

I hooked it up to a scope and could at least see that there was a bunch of communication on MOSI with SCK giving the accompanying clock signal, but MISO remained completely silent.

>>1173504
Do you have a pull up on the reset pin?

File: IMG_20170507_135059.jpg (1MB, 1560x2080px) Image search: [Google]

1MB, 1560x2080px

Can /ohm/ elucidate for me these symbols?they seem like caps of some description but I don't recognise them

>>1173493
>>1173506
Top kek
All this time I had mosi and miso reversed like a retard.
Well now I can read the fuse bits at least, low is 0x3A and high is 0xFD
Still won't let me program (or "burn bootloader") from the Arduino IDE with MicroCore though.

>>1173508
Neon lamp and a non-polarized (electrolytic) capacitor.

>>1173512
Interesting. Uploading from the Arduino IDE didn't work for shit, but using the standalone (latest) avrdude to manually upload the compiled binary to the tiny13 worked like a charm.

At least that's confirmation the programmer and the chip work correctly. Now to wrestle the IDE into submission.

File: 2017-05-07-153854_1366x753_scrot.png (46KB, 1366x753px) Image search: [Google]

46KB, 1366x753px

>>1173512
>>1173518
Just tried mine with MicroCore.
Works fine for me.

Is there a component which has 4 diodes in a single shell and they should be individual not connected up in rectifying topology? If so, what english word I need to use to search for it from distributors?

>>1173530
*by diodes I mean 1N4007 or equivalent.

File: WhatsApp Image 2017-05-07 at 16.46.43.jpg (96KB, 747x1328px) Image search: [Google]

96KB, 747x1328px

hey /ohm/

I'm using a relay to turn a transformer (in pic related) on and off. I switched to a solid state relay (opto 22 240d10 - 240 VAC, 10 Amp, DC Control) instead of a normal one since the normal one got stuck now and then.

But since I started using the solid state relay the transformer sometimes stops working. Then I need to press the breaker switch / reset switch on the bottom and I can use it again.

Does anyone know what is wrong and how to solve it?

>>1173530
Diode array, but I don't remember ever seeing anything but signal diodes or schottkys.

>>1173560
Cheers. Used that keyword for search and found out there actually was some component now obsolete and twice larger than my need.
Guess I stick to individiual diode component for now.

>>1173560
There's plenty:
https://www.digikey.com/products/en/discrete-semiconductor-products/diodes-rectifiers-arrays/286

>>1173569
Found this S8-4150/TR13-ND from your link. Everything about it is perfect for me, but not minimum quantity, "not stocked" option and price ;(

Thanks anyway.

>>1173492
I see. I'll probably glue a sheet of aluminum foil at the bottom to reflect the UV light that bounces back from the glass, but I'll keep the walls black, since light bouncing from the walls could have a very inclined angle and could get in between the exposing film and the board.
I'll leave around 4 to 5 cm between the LEDs and the glass to ensure a good enough diffusion of the light.
The LEDs should work fine as long as they are true UV with a wavelength below 400 nm and not a mix of other colors.
I added intensity as a parameter just in case the light is too strong and causes problems when exposing. I guess 12W on each side should be enough.

>>1173585
(different anon here)
I'm not quite sure what you mean when you say you will add a layer of reflector "bottom" to glass, but unless you are using some special type of glass like borosilicate the uv will all be not transmitted through the glass so adding a reflector beneath it will bu utterly useless.

is it safe to buy PSUs, multimeters and oscilloscopes from China (Alibaba)?

>>1173606
no

>>1173606
no, yes, technically but don't do it

honestly you probably have a few fire hazard power supplies anyway if you've bought random shit off there before so don't freak out about it. if you know smps basics you can open some of them up and decide for yourself.

File: 9000.png (269KB, 1200x800px) Image search: [Google]

269KB, 1200x800px

>>1173596
>uv will all be not transmitted through the glass
Wikipedia:
>Ordinary window glass passes about 90% of the light above 350 nm, but blocks over 90% of the light below 300 nm.

The strip I bough emits light from 395-405nm, just in the beginning of the UV spectrum. I will lose a 10% to the glass, but that's fine. My only concern is if the LEDs are enough to expose the photoresist, I've seen other people use UV LEDs and glass, so I guess it will.

I have a cheap shitty projector I want to turn into a cheap "touch screen" of sorts. I'm not looking for anything super accurate or fancy for actual detection, just something easy on a microcontroller. I've already got vga running for video but I've been thinking about how I'll actually detect touch.

I've seen two methods that look pretty feasible, one using a trio of piezos to sort of triangulate the touch, the other using IR light and sensing when paths are disrupted.

I think I'm going to go for the prior but I just was wondering on any input in case I'm missing some way obvious

>>1173617
Ah yes, sorry, you were talking about UV from another region that I am been using. You use UV-A region while I automatically was thinking of thing like UV-C. Sorry. My mistake.

But if I may make suggestion: replace glass rester plate for PCB with clamps holding from two sides so no loss from both sides of radiation and bring UV lights closer to PCB because intensity loss at distance grow exponential? Only suggestion, you do what you decide is ok for your design

can you recomed me a nice H-Bridge motor driver and a comperator? Preferably THT parts.
Motor driver needs to handle 10V 1A. Controlled by 5V TTL.
Comperator needs to handle 0-10V I want to hook it up to a microcontroller pin via a voltage devider so it doesn't need to be capable to drive much current.
Multiples in one package are welcome.

>>1173631
BA6956 is cheap and in an adorable outdated sip package. it might not handle PWM well above the low khz.

i've used TLE5206 before but it's overkill.

>>1173639
by the way what uc are you using? a lot have a built in comparator. i know you said 10v but you can almost certainly modify your circuit to work with the 3-5v limit on a uc comp.

>>1173631
A SN754410 can handle that

>>1173651
AT89S4051 It has one comperator, but I need 4.

>>1173631
>>1173639
>>1173651
>>1173655
>>1173664
where can I learn about this stuff?

>>1173666
depending on how new you are, you should start by reading a book like the one pictured in the op. you don't need to read all of it. just enough to understand the fundamentals. you can fuck around in multisim or another simulator to reaffirm that knowledge.

hopefully while reading you'll find more specific subjects that interest you. for me that was power electronics so i started reading about that, which itself branches into subjects like transformers and switchmode converters and control loops that are all their own little worlds.

somewhere in there you pick up ideas for circuits you want to build and you buy a bunch of cheap parts off ebay. when you have those circuit ideas you search around on how to achieve them or browse distributor websites and find parts like the ones mentioned above. then you remember them when someone else asks.

>>1173681
i'm an electromechanical engineer, but i don't know much detail about the electronics driving a motor. I only had one electronic circuits class where we learned basic shit like nodal/mesh analysis, flip flops and transistor/opamps (general working principles).

>>1173685
i'm mechanical, my interest in electronics was sparked by my senior project and the rest is self learned. in my opinion university doesn't teach very much that's practically useful anyway.

as far as project scale motors there's not a huge variety. there's servos, unipolar/bipolar steppers, brushed, and brushless motors with/without hall sensors. you can find driver circuits for each of them online that don't change much between implementations. integrated solutions can be found on digikey under the PMIC categories.

>>1171979
probably fucked up because i already asked in the rc general

i'm a casual to imageboards but remembered this place, i hope you can help me with this issue.
I designed a gps patch antenna, the deal is that i dont know how to feed it, by this i mean , can i just take an rg147 cable take the inner and outer conductors and solder them to the patch and gnd plane, or do i need fancy sma connectors between the antenna and the cable? thanks.
i think some of you may have practical experience with antennas and stuff .

If I'm just doing it for ripple rejection, not as part of any tuned filter or anything, is there any reason I can't use a substantially bigger cap than an existing one?

>>1173753
No. The only cases I know of where increasing capacitor size can be a problem are ceramic bypassing where esl supposedly increases with capacitance and dc-dc output caps where larger sizes can destabilize the control loop.

>>1173736

Just solder it directly to the antenna.

Keep the stripped end as short as possible.

Just starting out with using arduino's, did a few projects but would like to try connecting to a simple HMI on my phone. Next project is a counter pressure bottle filler and I would like to be able to change the filling amount and adjust the filling speed. So basically passing a few values from phone to arduino.

I'm not married to the platform, just starting to become familiar with it. Hardware would consist of a hall effect flow sensor, 3 valves, diff pressure sensor, a few buttons, and a stepper motor.

Should I stick with the arduino and pick up a wifi shield? Or is there a board that would be better suited for this? There are a ton of choices out there.

>>1174015
You could look into cheap ESP8266 based boards like NodeMCU or Wemos.

The ESP8266 is what's often used in those Arduino wifi shields but they can also be easily programmed directly with the Arduino IDE. They're more powerful than the AVR on a normal Arduino but have less I/O (7 digital, 1 ADC AFAIK).

File: s-l300.jpg (12KB, 300x200px) Image search: [Google]

12KB, 300x200px

Has anybody used this? I want to tin my circuit boards and this and "liquid tin" are pretty much the only options, but I don't want to spend 23$ on a 125ml bottle of liquid tin.

The instructions call for dissolving it in lukewarm water and letting the boards sit for around 5-30 minutes.

It's much cheaper than liquid tin, but so far I haven't seen anybody using it.

>>1174067
URL:
http://www.ebay.com/itm/Preparation-for-Electroless-Copper-Tinning-PCB-Tin-Plating-Crystals-Immersion-/122452023828?hash=item1c82b59a14:g:BFEAAOSwjVVVhVQG

File: 1372998465072.jpg (20KB, 290x345px) Image search: [Google]

20KB, 290x345px

when you make 240v from two 120v hots, does the breaker amperage double or stay the same? ie if both legs are on 15a breakers (so 120v@15a x2) do they combine to 240v@30a or stay at 240v@15a?

>>1174067
it works, but the solution doesn't seem to last more than a use or two so i wound up just running my soldering iron with a little solder across all the copper on boards instead

>>1174393
you have a single 240v circuit with 15a breakers (one will probably fire before the other due to manufacturing differences) on both ends, so 240V15A max

>>1174410
k thanks

>>1174393
The size of the breaker should be determined by the gauge of the wire it's on.

>>1172413
https://www.amazon.com/s/ref=nb_sb_noss/146-7236767-4704517?url=search-alias%3Daps&field-keywords=electronics+kit

File: Untitled.png (13KB, 512x512px) Image search: [Google]

13KB, 512x512px

>>1174438
not sure how that applies?

File: parallel series.png (3KB, 415x333px) Image search: [Google]

3KB, 415x333px

>>1174393
>>1174410
Aren't 120V sources from breakers just all in parallel behind the breakers? If you've got isolation transformers or other seperate AC sources then that's fine, but trying to connect two parallel voltage sources in series to add the voltage just ends up shorting two contacts. And you need to make sure that the two AC sources are in phase. Ideally the AC coming out of a transformer would be in phase with the input, but in reality there is a shift and it can be as high as 30°. So either adding waveforms from two separate isolation transformers or two inverters with a shared 60Hz clock would be about the only way I could see to get it to work. You shouldn't have any problems with the 50Hz/60Hz inconsistency.

>>1174627
>>1174570
If you're in north america, neutral is zero volts, live is +/- 120VAC
neutral doesnt change in voltage, so this won't work just fyi

>>1174406
>i wound up just running my soldering iron with a little solder across all the copper on boards instead
I did that too, but it's no longer feasible with big boards and fine traces, plus it looks quite shitty.

I'll give it a try though. They claim the solution remains effective up to two months, as long as I can plate a couple of boards each time it'll be worth it.

>>1174570
The amps always determine the gauge of the wire. The insulation determines the volts.

>>1174669
It will if each circuit is hooked up to a different rail in the box like pictured.

>>1174063
i will do some reading thanks anon

File: Untitled.png (14KB, 512x512px) Image search: [Google]

14KB, 512x512px

>>1174627
in na each slot alternates between "0" and "180" which is why 240v double breakers are always one on top of the other

>>1174669
notice how the plugs are coming from 2 out of phase legs and the wires are coming from the hot leads of the plugs

>>1174681
im talking breaker amperage not wire amperage. this is for an already installed electrical system i need 240v from but not worth installing a socket for (plus its a rental anyway).

>>1174787

in case it isnt clear, you already got your answer: a 15A breaker will break at 15A whether your load is 1 volt, 120V or 240 volts.

>>1174790
in case it isnt clear, im aware of this and trying to explain to you what i was talking about but since youre just going to be a giant douche then go die in a fire

>>1174787
> amperage
A better word would be 'current'

>>1174791

just trying to set the record straight after the usual confusing and contradictory replies. your gratitude is overwhelming.

>>1174793
This. Fucking hate people who say amperage. Their opinion is immediately dismissed and they automatically look like retards.

File: 1374900635346.jpg (29KB, 396x400px) Image search: [Google]

29KB, 396x400px

>>1174793
>>1174802
>ugh why dont people use this word that means the exact same thing as the word they use?
>i know right like what a bunch of retards using a literal synonym for this other word hahaha im so alone

>>1174820
No. It's like when people say 'preggo' or 'hubby'. You just sound completely retarded, and that's just a fact.

>>1174820
>posting bill nye the science goy
>post sex junk

nope.

File: Chancelor Butthurt.jpg (20KB, 494x358px) Image search: [Google]

20KB, 494x358px

>>1174826

>>1174820
cut off your son's penis

So I am looking to make a capacitor bank because at some point in the future I want to play around with a little home made coil gun. My question is about capacitors as it is my first time working with them. For this application do I want high voltage (multiple 100+ V ones) and/or high charge (1000+ uF)? And does the type of capacitor matter too much in this application, like ceramic vs electrolytic?

>>1174900

since you need a lot of capacity to generate a lot ampacity, then electrolytics are the most cost effective way to go. supercaps are low-voltage but give the most bang for your buck.

>>1174904
So I should be looking for high capacitance (10+ F) rather than voltage? I plan to continue learning about them before they arrive and then attach them in series and hopefully even add a small circuit of leds to show when they are charging or ready for discharge.

>>1174904
>ampacity
into the trash it goes

>>1174900
You can calculate the stored energy with W = C/2 * U^2.

But you also have to mind the ESR. For coil guns you need a high current so a low ESR becomes really important. Supercaps tend to have higher ESR so they're not suited for it.

You should look at Photoflash capacitors. They are specifically designed to give short high current bursts.

File: capacitor_energy.png (16KB, 300x225px) Image search: [Google]

16KB, 300x225px

>>1174900
Use common sense. What have they taught you about capacitors in class? Capacitors store energy, the energy stored in them depends on capacitance and voltage, the energy increases linearly with capacitance, but exponentially with voltage, this means increasing your voltage is more desirable. It's like filling a compressed air tank, if you want to fire a projectile out of an air gun you don't want a lot of air, you want a small amount of air with a high pressure.
You want your capacitor to have around 200 to 400 volts, higher voltages are stupidly dangerous and can kill you in an instant.

If you look for capacitors you'll see the highest energy density belongs to supercaps, but those operate on low voltage, the next ones are electrolytic capacitors, which are ideal for a coilgun. Ceramic capacitors, although they're good at handling very high voltages, have a poor energy density.

>>1174907
>10F
No, unless you want your coils to melt into a puddle of molten copper. The voltage must be high to push those electrons though the coil and the capacitance should be relatively low to create only a short but very strong electromagnetic pulse to fire the projectile.

>>1174907
for my coil gun project I used 35 volt rated 4700uF capacitors for the bank.

File: Screen Shot 2017-04-29 at 6.15.15 PM.png (169KB, 1476x770px) Image search: [Google]

169KB, 1476x770px

Supercapacitors technically give you better energy-stored per/dollar (at least going off Jaycar prices), but one has to take into account that they can provide significantly less current because of higher internal resistance. I'd recommend taking a few different capacitor stats and throwing them together with the total energy you need (100J-1kJ if you're sensible) along with some current calculations and guessed inefficiency factors and you should get a good idea of which caps to pick. Note that if you pick caps that have a high voltage rating and low current capacity you'll want to wind more turns than if you had the same energy but with lower voltage and more current capability. Have fun!

>>1174910
>higher voltages are stupidly dangerous and can kill you in an instant.
safe is boring

>>1175133
Get a load of this faggot
I'll be sure to laugh at your obituary you amateur

>>1174802
>>1174793
Amperage is to amperes as voltage is to volts. I know you probably pretending to be retarded but voltage and amperage refers to a value being provided by the circuit (nominally).

>>1174904
Ironically, despite using ampacity correctly (an ability to provide amperage), most storage provides better ampacity the worse capacity it has per unit comparison i.e. a three volt battery with half the Ah capacity would provide twice the amperes than a 1.5V with the 'whole' Ah capacity simply because it would deliver half the current for a given load. There's a reason we use like 500 3.7V batteries in electric vehicles rather than 1 giant 3.7V battery.

>>1175181
>Current is to amperes as voltage is to volts
Yes, we can all agree on that

>>1175181
Amperage is what shitlords call current, you are the retard. Technically Amperage specifically refers to current measured in amps, but if you're not both measuring current in an SI derivative of the Ampere and providing your value with relevant measurement dimensions then you are a shitlord.

Also Ampacity doesn't mean shit when dealing with pulsed currents, it only refers to continuous loads. You want fusing current, and should do some heat dissipation and insulation calculations. Ampacity is an ok ballpark figure, but I wouldn't worry about going a few times over that in a coilgun assuming you've got it configured to both turn on and off very quickly.

Also also, the 500 3.7Vs in EVs are half in series, so their total current capacity remains at shit-all. They do this to get higher voltage. There would be nothing wrong with using 50 or so massive Li-ions in series to get the same voltage and current capacity, but they don't do it because replacing a single faulty 18650 or whatever they use is far cheaper than replacing a 20kAh Lipo. Two cells in parallel is equivalent to a single larger cell.

>>1175181
>amperage
>ampacity
I wouldn't take seriously the ideas and opinions of someone who uses those as legit terms

>>1175181
I thought voltage was the generic English term, not just the measured value in Volts.
In my mother tongue and many other languages, you literally say tension/pressure. I've never seen it called that in English.

>>1175211
I've seen tension but more informally. Also potential difference. Voltage is definitely the most commonly used words though, on scientific texts too (especially in engineering).

Amperage on the other hand is universally agreed upon as being part of the language of idiots.

My mouse just died. I think it'd be fun to try to fix it since I know nothing about electronics.
Buttons and scrollwheel still work and I can make the mouse move by moving a flashlight in front of the sensor.
1) I think the led died. Could there be another culprit?
2) The led my mouse use seems to be an infrared one. Can I replace it with any other kind of led or should I stick with infrared? (I have blue, red, green and yellow leds lying around)
3) What's a good soldering iron for a beginner? I plan on doing more than just fixing my mouse (e.g. building an ergodox keyboard this summer).

>>1175263
Do you have a multimeter (any cheap piece of shit will do here)?
Then you can check the voltage drop (around 1.5v for an IR LED) on the led and the current flowing through it (somewhere around 5-20mA) while the mouse is plugged in.

>>1175278
I don't have a multimeter but if I put a led on the dead led it turns on.
What's the purpose of checking the voltage drop? Finding out the kind of led I should use to replace it?

>>1175279
If you measure current but no voltage drop on the led, it means the led is broken.

>>1175279
if it turns on when touched maybe you've got a cracked or corroded trace/pad.

>>1175281
nvm misread your post

>>1175280
I understand. Thanks!

>>1175281
I meant the other led turns on. But that's normal I suppose. Thanks for the input anyway.

>>1175279
>if I put a led on the dead led it turns on.
IR leds have a drop around 1.5V, whereas visible ones are 2.2V and up. so, if you put an IR and visible LED in parallel, the visible wont light coz the IR one drops the voltage below it's turn on point.

long story short: your IR led is burnt out. i doubt you can pull one off of an old remote control to replace it, coz it wont be rated for so much current.

thrift stores carry mice for like $2-$3, so i probably have 8 replacements on hand.

File: vtsc30n.jpg (32KB, 350x263px) Image search: [Google]

32KB, 350x263px

>>1175263
Check it with your phone, IR LEDs are visible to cameras. If it doesn't shows a purple light order a bunch for a couple of cents and change it.
I would start with a velleman sooldering iron or other cheap soldering iron which doesn't has a flimsy tip, those are rubbish. Pic related lasted me about 4 years.

>>1175329
Thanks for the suggestion Anon. I plugged it back in and it seems to flicker. I have no idea why this could be happening.
And thanks a lot for the recommendation.

File: 200px-Neckbeards.jpg (29KB, 600x585px) Image search: [Google]

29KB, 600x585px

>>1175185
>>1175201

>>1175201
Ampacity is a perfectly legitimate term, just look at its wikipedia page. Amperage does however sound somewhat unprofessional.

>>1172866
I crimped it back and it appears to be ok. However, stripping the insulation revealed blackish powder, probably oxide of the copper wire. This appears to be the failure point, as it increases resistance, which in turn creates heat, which in turn creates more copper oxide. The stranded copper wire is brittle all through because it's been oxidized so much

>>1175354
>>1175419
I didn't know /diy/ had such a high proportion of rubes and bumpkins

new guy here. I'm studying EE and can't get a consensus on why AC needs RMS vs. average voltage or current.

almost everyone on stackexchange and the electronic forums say that it's simply because when you average a sinusoid it averages to zero.

but occasionally there is one guy out of 100 who says it's not that at all, it's because the average of squares is not the same as the square of averages.

I am very confused. Can you help me?

>>1175920
>the average of squares is not the same as the square of averages.
((x + y)/2)^2 = (x+y)^2/4 = (x^2 + 2xy +y^2) / 4
!= (x^2 + y^2) / 2

>>1175920

it's real simple. RMS is the *process* by which you calculate the average of a sine wave. so RMS = average.

(if you try to simply mathematically average something which is positive half the time and negative half the time, the obvious answer is 0, which is wrong, and that's why you need a diff method of calculation, like RMS)

>>1175925
Then why not just take the average of the absolute values?

>>1175929

it would give you the same answer. but the acronym wouldnt be so funky.

>>1175925
No. For sines, the RMS value is 1/sqrt(2) * amplitude, while the average of the absolute value is 2/pi * amplitude.

hey
I'm building a dual rail psu for a function generator. It consists of a +-16V iron core transformer, filter cap bank and the pair of 7812 and 7912, so I want +-12V regulated output. The thing is, at first 7812 output 12 volts and 7912 outted 16 volts. This got resolved by adding a resistor so at least 5mA flows all the time. Now 7812 outs 12.15V and 7912 does 11.97V. My real question is, would this ~.2V difference cause problems? Already tried it, it works, but I'm afraid it will fuck it up in the future. Also, any idea what would cause this? The rails are perfectly symmetrical, so I've no idea.

>>1175942
>Also, any idea what would cause this?
The nominal accuracy of the 78xx and 79xx regulators is around 4%.
If your function generator is unhappy with your positive rail's 1% error, then it's shit.

>>1175945
thanks, didn't know that. It will be okay then, since it works flawlessly, I just tend to worry about stuff like this.

File: 555 mono.jpg (15KB, 887x591px) Image search: [Google]

15KB, 887x591px

What would be the best way to activate a relay with this circuit?

I'm not good with stuff like this, but I've tried to replace the LED with a wire to activate my relay but it won't do anything...

File: IMG_20170511_193121.jpg (1MB, 2221x2238px) Image search: [Google]

1MB, 2221x2238px

Any anons know what this is, can't find any markings. Kinda newfag so please use english.

>>1175976
Ferrite rod antenna, also called the loopstick antenna. Commonly used as an antenna in medium wave (AM) radios.

>>1175973
>I've tried to replace the LED with a wire to activate my relay but it won't do anything...

presumably you mean 2 wires, plus and minus.

anyway...

1) make sure you have a reversed diode across the relay coz when they release, they can generate a huge voltage spike that can kill girly chips like a 555. (in fact it may already be dead - test it with the LED again)

2) get rid of the 470 ohm resistor, of course.

3) the voltage powering the 555 must be the same as the relay's rated voltage.

4) the relay should not take more current than the 555 can give. from memory, the max is 0.2A. might be smaller for the fruity CMOS versions of the chip.

5) there was another important consideration but i forgot all about after downloading the datasheet. so, make sure that's taken care of as well, whatever it was.

>>1175973
well typical 555 can sink/source 200mA of current.
probably not enough to drive a typical relay. maybe a solid state relay or optoisolator.
the normal way is to drive a transistor. typically npn, make sure collector current can drive the relay and collector current you need / hfe is less than the 200ma output from the 555. don't forget freewheel diode for emf from relay field collapse

>>1175994
Coming to think of it I wouldn't need a relay. I need it to control a latch on a room door. It runs on 12v so just skip the relay and have the latch power hooked up to the output. In theory, that should work right?

>>1175999

not likely. a latch is gonna take about 0.8A, which is 4x what the 555 can give. but you can use a power transistor or a FET to multiply the current output.

>>1176001
I'm looking at the specs of the latch and we're looking at 12VDC 0.1A.

>>1176005

cool, but 1.2W is pretty damn weak for a latch. anyway, you still need a reverse diode across the latch, for the same reason as the relay.

File: relay-circuit.gif (3KB, 372x314px) Image search: [Google]

3KB, 372x314px

>>1175973
Add this to your output

>>1175929
>>1175932
The answer would not be the same. The average of the absolute value for a sinusoid is 2/pi*Vpp or around about 0.64*Vpp. By taking the mean of the square of the waveform you are averaging its power as opposed to its voltage, and this gives us the √2 that we're used to.

The equation for an arbitrary waveform f that varies upon t for its RMS value is √(∫V(t)^2•dt), with the integral definite from 0 up to the wave's period T.

I'm a physics student by the way.

>>1176038
How are you averaging its power without a specified current?

File: Screen Shot 2017-05-12 at 5.13.47 PM.png (208KB, 766x662px) Image search: [Google]

208KB, 766x662px

>>1176039
You can derive it using the V=IR and P=V^2/R for your waveform, and with a little integration, the Rs cancel out and the numbers all work out in the end, probably.

I'm considering buying a weeb iron (pic related) since my shitty $30 budget iron died on me after 3 years. That's 65.60 USD btw. How bad of a deal is this, and is the brand worth it? Is 80W (it might actually be 85W) too much? Any information is welcome.

On the flip side I could get a $35 10W temperature controllable soldering station, I don't know how to feel.

>>1176114
Looking at the Goot website it seems the highest a single-temperature iron goes up to is 38W, and only 30W for one that looks exactly like the pic. The 30W that looks like it goes for $43 US on amazon and the 38W doesn't exist anywhere. I get the feeling I'm falling for a meme here.

>>1176039
RMS is the equivalent power into a resistive load, hence the assumption that P=V^2/R.

>>1176119
Don't see how this would make sense as a meme, there are other soldering pencils like the weller wsp80.

>>1175923

correct

>>1175925

wrong, like almost everyone else on the internet.

forget sine waves for a moment, so that the point of RMS can be made clear.

consider a pulse that is 10 volts for 1 msec then off for 9 msec on a 1 ohm resistor.

clearly the AVERAGE POWER is 10*10/1 for 1 msec = 100 watts for 1 msec, but divided by 10 for the whole 10 msec, so it is 10 watts.

if i use AVERAGE VOLTAGE then i get 10/10 = 1 = AVERAGE VOLTAGE which says the average power is 1 watt, which is off by a factor of 10.

The whole confusion of RMS (root-mean-square) seems to be the R part. the MS part simply says take the AVERAGE OF THE SQUARE OF THE VOLTAGE FUNCTION.

then we throw in the R = root so that we are back to volts (or amps), which we then SQUARE to calculate power in a resistive load.

>>1175929

wrong

>>1175932
wrong. stop being clever when you are clueless.

>>1176038
correct

>>1176127
correct again

>>1176131
When a board brags about something so much that it's hard to tell if they're trolling or not, "Falling for the meme" means when you take their advice as real when they were actually trolling. Meme does not simply refer to a humorous reference, but to any popular aspect of culture. Thinkpads and Gentoo of /g/ are an example, so are shipping containers over here.

File: RMS sinusoid.png (4MB, 2000x2000px) Image search: [Google]

4MB, 2000x2000px

>>1176174
Take my 3.7MB.

>>1175920
This chapter explains the point of RMS
http://www.ibiblio.org/kuphaldt/electricCircuits/AC/AC_1.html

In short:
>.. is a way of expressing an AC quantity of voltage or current in terms functionally equivalent to DC. For example, 10 volts AC RMS is the amount of voltage that would produce the same amount of heat dissipation across a resistor of given value as a 10 volt DC power supply. Also known as the “equivalent” or “DC equivalent” value of an AC voltage or current.
>Unfortunately, an AC measurement based on work performed by a waveform is not the same as that waveform's “average” value, because the power dissipated by a given load (work performed per unit time) is not directly proportional to the magnitude of either the voltage or current impressed upon it. Rather, power is proportional to the square of the voltage or current applied to a resistance (P = E2/R, and P = I2R).

>>1176017
A capacitor between collector and emitter can't hurt either.

>>1176204
Let's take a practical example and say we have a 20Ω resistor with an AC voltage of 120Vp across it, and figure out what the equivalent DC voltage would be for the same power consumption.

Our voltage waveform will look like V(t)=sin(wt). Since P=V(t)^2/R, we can see that the maximum power dissipation will be 720W, so we'll get a sinusoidal power usage of P=720*sin(wt)^2, which through trigonometric identities or some shit we can equate to P=360-360*cos(wt). By integrating this over time (sum of all minuscule sections of E=t*P, because P=dE/dt) to get the total energy spent and dividing it by the time elapsed over some multiple of periods, we'll get the average power dissipated. Some tedious shit later, and we get E=[360*t-(360/w)*sin(wt)] across (0, 2π/w). Taking the final-initial for E(t), or E(2π/w)-E(0), we get E=(360*(2π/w)-(360/w)*sin(w*(2π/w)))-(360*0-(360/w)*sin(w*0)). Simplifying, and using sin(0)=sin(2π)=0, we get E=(720π/w), which is neat. Glad you're not using an out of phase sinusoid now, aren't you? Divide this energy by the period of the wave to get the average power and we get P(avg)=(720π/w)/(2π/w), or P(avg)=360W. As P=V^2/R, V=√(PR), so V(rms)=√(360*20), or V(rms)=84.85V. This equals 120/√2, or in other words, V(rms) = Vp/√2.

The reason we use this method is because when comparing AC voltages to DC voltages, the only really thing that matters when using these voltages practically is how much power they dissipate on a load. You would probably want to use the peak voltage when deciding the needed breakdown characteristics of the used components, and average voltage when determining anything that scales directly off voltage, such as the average charge within a capacitor. Note that the voltage across the capacitor would have to be rectified to be DC, otherwise it will simply dissipate power across the capacitor's reactance, for which an RMS value would again be needed.

File: psu.png (8KB, 524x481px) Image search: [Google]

8KB, 524x481px

I'm using a boost converter to create a high voltage DC supply for a project I need but I also need a 12V rail to power some logic circuitry. Can I just use a simple voltage divider from the 180V rail plus a linear voltage regulator to give me my 12V? The load on the 12V rail should be relatively small, just a small class A amplifier some MFB bandpass filters, and precision rectifiers, nothing that will be pulling a large amount of current.

File: 59265230.jpg (67KB, 864x717px) Image search: [Google]

67KB, 864x717px

How many volts does it take to be able to feel it?
I was working with 20vdc today, wasn't sure how cautious to be.

For example, dry skin, touching + and -, about 10cm apart on the forearm?

>>1176243
In principle yes, but that divider in your pic is far too wimpy to even supply the regulator's own quiescent current. Also, the current dimensioning provides only barely sufficient voltage for the regulator and any loading makes the divider voltage to drop below the regulator's minimum input voltage.

>>1176244
up to 24VDC is considered safe according to some standard I forgot the name of which is used for industrial automation.

>>1176251
Yeah never mind I'm retarded. Reading up on the datasheet for the 7812 and there's no way to do this without burning up loads of power in the resistors. I'll just have to figure something else out.

>>1176244
Watch ElectroBoom's video's on the topic, they're very practical and informative. As far as I remember, you can put your fingers across rectified US mains voltage (~170VDC or something) and not hurt yourself (much). AC is very different and starts hurting somewhere between 20VAC and 80VAC I think. Most painful frequency is 1kHz. Don't let electricity pass through your heart. Capacitance of the body. Also note that the distance between parts of your body doesn't make much of a difference to the hurt.

>>1176257
You probably have to use something with an inductor/transformer in it, a FET or too to get the AC. Dirtiest way possible is probably using a reverse bias NPN as a relaxation oscillator that you use to power a FET (or beefy BJT) and drive a small transformer that you might want to hand-wind, with a little rectification on the other end. Throw that into a sufficiently rated 3-terminal regulator and you're golden. But considering you're already using a boost converter to get the 180V you should probably just buy/make another one of those, or at least share half of that converter's circuitry with the 12V converter. You're also probably not a shitlord, so there's that.

>>1176261
>ElectroBoom
As a heads up, sometimes he pretends he's getting shocked by mains while he's really just touching a capacitor, so, unless he specifically says it's safe, be careful about replicating what he does.

File: zener regulator.gif (7KB, 364x242px) Image search: [Google]

7KB, 364x242px

>>1176243

this is the simplest way to do what you want. you would probably want the zener to conduct (at least) twice as much current as the load, so do your calculations accordingly.

i've seen this exact circuit used to generate 5Vdc from 120Vac inside a commercial product. for good form, you'd add a fat electrolytic cap across the zener. at least 1uF per mA of expected load current.

File: LM386Pinout[1].jpg (9KB, 300x227px) Image search: [Google]

9KB, 300x227px

Hey guys I bought 5 pack of lm386n and apparently pin 1 and 8 can amp the voltage with a resistor if you dont have adjustable knob, but is there any other way of amping the voltage? I don't have proper resistor to amplify the voltage.

I am trying to amplify the read from magnetic head in arduino.

Can someone explain wye configuration to me? I get it. 240v line to line. 120v line to neutral on A and C. 208v line to neutral on B

How do you get 208v? I understand mathematically, square root of 3 etc, but from where is it drawing the extra voltage and why?

Is it because of the distance crossed over the resistors in relation to the neutral in the transformer?

>>1176261

That that asked that^ question here. I'm an electricians apprentice, I've been hit by 110v a good handful of times. If you only get hit on one Hand and your standing on a fiberglass ladder and have good boots, you it's nothing. Across two hands on the floor in gyms shoes its a decent jolt, but still nothing really

The real problem is it makes you twitchy the rest of the day

In a simple audio synthesis circuit, is a voltage-controlled resistor the sensible way to have LFO control things?

i got comfy static dissipative boots free from work. what are the properties that make them work?

>>1176448
being conductive?

>>1176472
but then what's the difference between wearing them and just being barefoot

>>1176474
The difference is you wear shoes.

Also they have a megaohm resistor between you for safety.

How come no bigclive in the OP?

>>1176492
bigclive is an anti-brexit shill

>>1176498
This (if even true) makes his entertaining explanation of real world circuits invalid how?

(((>>1176500)))

I've got an L298N dualH bridge hooked up to a pwm-source and a motor, and it works great between 0-90%, but above that the motor starts going "choppy", like it'll cut out randomly for a tenth of a second or so once or twice per second.
What might I be doing wrong? I have the motor hooked up to Out1 and Out2, with the pwm source to In1. I'm running it on a 2S LiPo, so voltage is roughly 7.4V.

>>1176492
Pretty sure the OP is out of room, hence link to last thread being in a new post.

>>1176524
what's your pwm frequency and how big is the motor

>>1176492
>How come no bigclive in the OP?

a typical bigclive video has him (1) buy some stupid shit at the dollar store, (2) make some minor and useless adjustment to it, and (3) call it a DIY project. he's the Ahmed Mohamed of DIY. in wikipedia terms, he is simply ''unnotable". kindly stop the shilling.

>>1176573
Most of his videos are going through the inner working of some cheap dollar store item, not modifying it.

>>1176267
>Replicating anything a poo in loo does

>>1176624
He's a sandnigger, not a pajeet.

File: IMG_0936.jpg (206KB, 3264x523px) Image search: [Google]

206KB, 3264x523px

I bought a 40W Goot iron for 30NZD, and I've no idea what these cable markings mean. I just want to know what temperature it can take.

I want to make my own high-lumen headlamp. To make it light, I assume I want to use a COB (but which cob, and where) mounted into something, and then have a battery pack for some protected 18650s (cannot remember the actual name for those) or AAAs on the backside, and then I will put it together with some sort of elastic band.

How the fuck do I make the circuitry to run the COB, especially with a draining battery? Do I have to make some sort of cutoff myself?

>>1176661
Also, do they make multicolor COBs? I would like to have an option to have just red light.

File: Dimmable Headlight WEKSI 120m 160 Lumens 4 Modes LED Headlamp.jpg (10KB, 300x300px) Image search: [Google]

10KB, 300x300px

>>1176661

whats the point when you can get everything you want for $7, without having to re-invent the wheel?

>>1176642
The iron started smoking as soon as I plugged it in, is this normal at all? The smoke was coming from the internals, not from the tip. I'll take it apart once it's cooled down; I do NOT want to set off my building's smoke alarm.

>>1176440
en.wikipedia.org/wiki/High_leg_delta

When dealing with AC, quantities (voltage, current, resistance) are vectors. The difference between two voltages depends not only upon their amplitudes, but also their phases. 208V is the magnitude of the vector.

>>1176529
The motor is unmarked, it's a standard small 9V thing. The kind with the round sides, flat top, and plastic back.
My pwm-source ranges from 0% to 100% at 500Hz.

>>1176685
increase the frequency to somewhere between 5 and 25khz

File: transformer_family.jpg (40KB, 647x410px) Image search: [Google]

40KB, 647x410px

>>1176681
whats the voltage of your goot? 110-120?
what about the voltage of your country?
210 - 220/240? now you see the problem.

You need to plug your goot into a traffo.
Building a traffo would be another activity for you, but hey, it's all part of /ohm/ fun!

>>1176642
For the same amount of energy 110V requires more current, hence thicker wires. 230V requires better isolation. In some rare situations 220V might be more dangerous to touch.

>>1176707
I set it to 9kHz and it works now. Thank you, anon.

I want to make a walk-through metal detector. My plan is to make a wire loop around a door frame, which will serve as an inductor. When metal enters the loop, the inductance changes. I'm going to set it up with some capacitors to make an LC oscillator. The change in inductance will change the frequency of the oscillator.

From there, I haven't put a lot of thought into it. If it's slow enough, maybe I'll just ADC it and examine the frequency in a microcontroller. Or maybe I'll mix it with the typical resonant frequency and look for beats.

What do you think? Feasible?

File: kicad.png (396KB, 1920x1080px) Image search: [Google]

396KB, 1920x1080px

Cad noob here

what kind of a pin is ground in Kicad?

Options:
> power output
> power input
> open collector
> open emitter
> passive

also, would pins one and two be open collector/emitter?

File: Screenshot from 2017-05-13 11:49:50.png (11KB, 996x119px) Image search: [Google]

11KB, 996x119px

does this mean that the chip considers Vss (which I'm told is the same as ground) as logic high?

>mouser.be
>10 2N2222 transistors for 17€
>17€
>10 transistors
W-What...?

>>1176816

>tfw you live 350 miles from digikey HQ

>>1176778
i would expect a power input, the level is being actively driven to 0v rather than floating or doing the driving.

>>1176808
yes, this chip is living in opposite land.
Vss is whatever you supply to the Vss pin, Vdd is is whatever you supply to the Vdd pin. Normally Vdd >> Vss and Vss is 0v however remember that voltage is all relative and stop fucking saying ground when you mean 0v they are not the same thing at all just because occasionally they share potential.

e.g. say your chip takes max 5v, you could easily supply 5v to Vss and 10v to Vdd by moving your ground reference. the chip no longer considers ground to be logic high.

>>1176808
Oh you and your shitty rhythm IC relic. You know, you aren't helping yourself by withholding information.
But yes, it does.

>>1176816
Steel cans are expensive. Get ones in a plastic package, like PN2222.

>>1176827
still though, on allelectronics it was 5€ for 10

>>1176827

extra info just confuses me

File: wat.png (71KB, 697x725px) Image search: [Google]

71KB, 697x725px

Can anyone please explain what the fuck is the "Register Clock" (RCLK) on this 8-bit shifter?
I sort of understand how 8-bit shifters work, but for the life of me I can't figure out why two clock inputs are required. In all code examples I found online, the RCLK is left L when data is inputted, and then just raised back up to H when that's done. I don't see the logic in that!

>>1176905
it's a buffered shift register. basically it's a shift register with a latch in front of each input. the benefit is that you can flip all the bits on simultaneously when you're done clocking by tying RCLK to QH'. that prevents weird logic that could happen as the bits travel along the shift outputs.

>>1176910
uhhh..... I didn't really get that.
Sorry, I'm really new to all this. What do you mean "flip all the bits on simultaneously"? Why would you tie it to QH'? What's a "latch"..?

>>1176905
say you have a shift register with two output pins on the parallel port. Qb is connected to an ice cream machine. Qa is connected to a nuclear bomb.
so how do you get the delicious ice cream without triggering the bomb? you need to shift the bit past the bomb output without the output actually turning high.
so thats what this does, you toggle the serial clock with the rclk held low or whatever until you are ready. none of the pins actually change although the data inside is changing. when you are ready you toggle rclk and the data inside the chip that you just shifted in is then pumped to the output pins all at once.

its like a boring old normal shift register but with a latch inside. what is a latch? also known as a flip-flop it basically just stores a value. probably a D type latch in there, go check that shit out.

>>1176914
in a shift register, when you change the clock line (SRCLK) to 1 it reads the data line (SER) and changes the first output (QA) to equal that. the next time you change the clock line to 1 it moves QA to QB before setting QA again to equal SER. so on and so forth. the inconvenience there is that when you're setting all the bits they have to "travel" along the register outputs to their final destination. which in some situations can result in weird logic that you have to do annoying things to circumvent.

a buffered shift register basically has its "outputs" kept hidden internally, and they're only actually applied to the output pins when you change the latch clock (RCLK) to 1. so when you've set all the bits you then do that and all of them appear on the outputs at once, preventing the issue mentioned above. QH' is an exception where it shows the hidden internal "output" of the last register. you can sacrifice the usefulness of that last output by tying it to RCLK and always setting it to be 1, so when it gets there it automatically triggers RCLK so you don't have to use another uc pin.

>>1176922
Oh wow, that's really simply and sounds very useful.
Thank you!

File: 2017-05-14-010933_891x1438_scrot.png (352KB, 891x1438px) Image search: [Google]

352KB, 891x1438px

>>1176816
lol
my local electronics store does this jewery too, sometimes I'd say "Guess it'll survive the nuclear war, then."

>>1174900
11.8V LiPo works better than 450V@7800uF for me. Use thicker wires tho.

>>1176955
wouldn't he need a really expensive battery for that? batteries' internal impedance or current supplying capability is insanely low compared to caps, unless you have something like a drill battery which can supply >80A. Also, the most important point, charging time.

>>1176678
>160 lumens
I was thinking more like 1000 lumens.

File: mims.jpg (129KB, 669x595px) Image search: [Google]

129KB, 669x595px

Shouldn't LEDs 2 and 4 always be on in this circuit?

I need to get a few little things I can jam multimeter leads in for a little test jig I'm making. What would that part be called? Not sure how to find that.
>>1176957
Power tool batteries are just 18650s btw.

>>1176969
i agree with you, actually led2 should be on than led1 if anything.
then again i am quite drunk.
get it spiced or build it and see!
of course if led 1 was not there then there wuldn'r be any PD over led2 when q1 was on but led1 being there there is the same pd across led1 as led2

>>1176979
banana sockets?

Where can I buy like 10 cheap stepper motors.

Step angle doesn't really matter to me, as long as its not like more 10 degrees per step

File: 1481147201821.png (260KB, 563x542px) Image search: [Google]

260KB, 563x542px

>>1176983
i just looked and found these. then i bought 5 of them and now it says there's only 5 left.

http://www.ebay.com/itm/172658257100

>>1176985
There are million of the same exact steppers on ebay and Ali

>>1176969
There is so much stupidity in this circuit, I don't even know where to start.

LED2 will clamp the voltage to the forward voltage of that LED and assuming that both LEDs are the same color, LED1 is barely going to illuminate even with Q1 deeply saturated.

Second problem: There is not going to be anything resembling a sharp switch action that they are claiming in the first circuit, because the NPN is above the LED. Assuming that there would be enough voltage to feed LED1 at all, you would get a somewhat linear increase in the current through it after going above knee voltage + Vbe.

What a joke.

>>1176729
I bought it at a physical shop, and I did double-check that it's rated for 240V as is my country's mains. I think the smoke was just some shmoo that got on one of the screws, after giving it a scrub the problem disappeared. After using it a bit it's a much nicer iron than my last one, mainly because of the thinner (and less oxidised) tip. But it's a shame that it doesn't fit in my soldering-iron holder. I find that the solder on the end of it oxidises fairly quickly, even though I'm using rosin-core solder. Is my temperature too high? I can always build a triac-based dimmer if so. Maybe I could try and mod it with a thermocouple since it isn't a fixed-temperature iron, there's room in the tip.

>>1176755
Or you could just put some high frequency AC across the inductor and measure the current; the reactance will change as the inductance changes. Might not be as detectable as an LC oscillator though.

File: Screen Shot 2017-05-14 at 12.46.26 PM.png (59KB, 908x500px) Image search: [Google]

59KB, 908x500px

Ok I built pic-related on a breadboard to try and combine my TRRS earphone mic signal with a music signal from my phone, but the music signal wasn't audible at all. The circuit has a buffer to reinforce the weak passive mic signal and a summing amplifier to combine the signals at the correct relative amplitudes. R1, R3, and R4 are pots in my circuit, and I noticed that attempting to change the gain by making R3 and R1 different also had no effect. Changing the 500k pot also did nothing. Should I be doing this to the bottom ring (ground) and not the sleeve (mic signal or whatever)?

>>1176755
you can wire it as an LC resonant circuit, feed it with a square wave tuned to its resonant frequency, and feed that into a peak detector measured by a comparator.

the thing is i don't know how much the coil's inductance will change with a small object like a knife so that might not produce a change that'll stand out through the noise.

>>1177058
Because X = wL-1/(wC), if you increase L by an order of magnitude and decrease C by the same, there will be a much larger difference between X at resonance and X when the inductance changes by a tiny bit, so you'll want to use the smallest capacitor possible. This means using the highest frequency possible, which will be your limiting factor. So first pick your frequency, then make the largest inductor you can, and then pick a matching capacitor.

>>1177074
you can get into the mH range with relatively few turns around a door sized area. that length of wire will have a parasitic capacitance on the order of 10-100pF even with generous spacing so this system would probably be stuck under 1 MHz. i'm too busy doing nothing to do more math than that but it's a neat topic to explore.

>>1177080
>L=µ*N^2*A/l
I never knew that inductance was directly proportional to area, I should pay more attention in electromagnetism lectures. Probably better to write it as L=µ*n*2*A*l where n is N/l. So with coil radius of 1m, 50 turns, "length" of solenoid 5cm, you'll get about 200mH, which is pretty damn good.

The (relative) permeability of iron is around 200, but since other metals have a permeability closer to air than to water/a human, you'll only be able to pick up iron or magnetic alloys with this thing. Say the knife you're trying to detect is 150mm long, 2mm wide, and 30mm deep, giving a volume of about 9000mm^3, or 9E-6m^2. The volume of the solenoid is 0.157m^2 or 1.57E8mm^2. 0.00573% of the solenoid's volume contains the knife, so by saying 99.99427%*1+0.00573%*200 we get the effective relative permeability of the solenoid wiht the knife. The answer is 1.0114. I've rounded the permeability of air down to 1, but it shouldn't make much of a difference.

Say you have picked 20MHz as your frequency, and by using a 200mH inductor, that leaves you with a FUCKING TINY capacitor. You could use a 1pF capacitor but I fear that the internal capacitance of the solenoid is significant at that level, so let's stick with a 1nF cap to be safe. The frequency for this would be in the realm of 10kHz. At this point you might as well be using a single coil of wire, which would get you somewhere between 1mH and 10mH. At normal resonance of the 200mH coil and 1nF cap, when the knife is introduced, the reactance increases to 161Ω. At resonance of the 1mH coil and 1nF cap, the reactance increases only to 1.8Ω, so higher inductance is definitely the ticket. All you need to do is detect the increased reactance somehow, maybe with some sort of bootleg high frequency AC comparator, though I guess at 10kHz you could just whip up something with an audio opamp. You'll also need to make it sensitive enough to not detect a person walking through, but that really won't be hard.

>>1177052
Update: I just figured out that when I plugged this thing in my computer went from TRRS mode to TRS mode and started using my build-in mic instead. Fuck do I need an oscilloscope. Even when I just have the two ends of the TRRS extension cable going to each other, it stays in TRS mode, got no clue what's going on, maybe not enough shielding/attenuating signal? And maybe I need to amplify the signal going both ways? I'm not really sure what the difference between an AC signal on one side of the opamp trying to go through compared to the other side, I thought it would just go through regardless because bakcwards voltage = negative voltage Input impedance might make a difference. Maybe it's because I'm using an inverting amp and something's expecting to receive a positive voltage, who knows. But since I just found out that my internal microphone gives better quality than my earphones, I might not bother.

I have a horn that plays sounds, La Cucharacha, and has a PA system. It's supposed to hook up to my car's fuse box but my car's a Honda and it's impossible to fit it in the engine. So, would it be possible to just hook it up to a 12-Volt battery?

>>1177132
Of course, but I'd give it its own fuse in case anything goes wrong in it, such as water ingress. An inline fuse should be fine, you can even get them for standard blade fuses.

File: A typical electret microphone.png (13KB, 220x190px) Image search: [Google]

13KB, 220x190px

>>1177114

reading your confused posts gives me a headache, but i can see right away from your schematic that you've wired the mic op-amp to have a gain of 1, which is nuts coz a typical mic only produces about a millivolt. you want a gain of 100 to 1000.
if the mic is an electret type it also needs a DC bias to make it work. see pic

as for the music signal, the 500K pot is really attenuating your signal. try 1k-5K instead.

as for where exactly your ground is, do trial and error. there are only 4 pins, so you should be able to find it in less than 136,854 tries.

>>1177148
I'm feeding the earphone mic from my TRRS headphones into the earphone mic receptacle TRRS socket on my computer. It already produces the right voltage so the gain should be 1. I just double checked that the DC resistance between the right and left channels is 30Ω or so and the resistance from the right channel to what I thought was ground is 15Ω or so which checks out, I'm more confused as to the nature of the signals sent forwards and back along the fourth, sleeve contact. Opamps (OPA2134) work at a gain of 1, right?

>>1177148
>>1177150
His power supply is fucked. The 100k ohms mean he can at most get 90 uA out of the entire circuit.

>>1177150
> the nature of the signals sent forwards and back along the fourth, sleeve contact.

there are no signals sent back and forth on the sleeve. it's ground, so it's just a reference point for the other 3 pins. you should read 15ohms between the sleeve and either speaker, which confirms that it is indeed ground.

the OPA2134 is a typical op-amp that can be wired to have a gain between 0 and a million.

>>1177152

not really true, since the op-amp takes almost no current from the artificial ground, all the supply current comes straight from the battery, however a better design choice would be to replace the 100Ks with 10Ks.

>>1176997
If LED1 and 3 are red and 2 and 4 are green, the circuit will literally work perfectly.

>somewhat linear increase
You really have no idea what you're talking about do you? If the red is ~1.8V, then there'd be a 2.5V at the base, and assuming a beta of 100 (low) you'd get up to 65mA through the LED - which you can't even get out of the circuit. Even 5V would permit more than 25mA. The other circuit would be biased at 2.5mA into the red, which permits the other 6.5 into the green, easily enough for the transistor to be in linear mode.

>>1177159
Look again. The op-amp isn't routed through the 100k, but there MUST be an ability for both inputs to pass current, unless you think his amplifier can literally run only taking 90uA max from the inputs. I also can't guess why he'd even have the capacitors there, since they're not even tied to ground.

>>1177165

huh? op-amp inputs take current in the nano-amp and pico-amp range. the OPAx134 named above is a FET input device that needs a mere 5pA for biasing the input pins.

>>1177167
That's not what that means. Those are inherent measurements related to noise and floors. With the 100k in line with the input, that 5pA is effectively a .5mV source. On top of that, electret microphones have built-in amplifiers, which is probably not working because of the huge impedance, and the only way it COULD get power is drawing from the op-amp.

File: Screen Shot 2017-05-14 at 8.43.45 PM.png (334KB, 2040x948px) Image search: [Google]

334KB, 2040x948px

>>1177158
It's the standard people use, not the shitty old Nokia one. I read 30Ω between the tip and first ring, and 15Ω between the tip and second ring, which is the ground under this standard.

>>1177172
Sounds like the electret microphone isn't getting the bias voltage it should, which makes sense since I could still momentarily short the mic and ground leads to act as a pause toggle button.

After a little testing, it seems that I had to unplug and replug the TRRS jack that enters the computer WITH the earphones plugged into the other side with the mic leads directly connected in order to get it to switch to external-microphone mode. But this trick still doesn't work when i'm plugged into the buffer+combination amp. This pretty much confirms that the bias voltage is what's missing.

Testing the bias voltage when the computer is in external mic mode gives me +1.17V. The electret mic pulls ~630µA standby, and ~590µA when the mic is being blown at. I'm not wondering how the computer decides how to switch between internal mic and external mic, not it seems I can do so by unplugging and replugging in the earphones to the extension cord, so I guess it isn't a switch or two in the socket but rather some electrical means. I really need a scope now, especially if I want to replicate this.

>>1177192
I don't suppose I can just bias the voltage divider a little to get the 1V?

On a side note, is there anything worth plugging into a cellphone that can take advantage of this possible >1mW power source? Was thinking maybe a radio of some sort, but I'm not sure. I guess that's probably how those TRRS geiger counter/thermometer/UV meter/EM meters work. Maybe I'll check to see what kind of power they can output.

>>1177194
With resistors between the computer-side mic-in line and the ground:
2.61V at 833kΩ (3.13µA)
2.59V at 338kΩ (7.66µA)
2.58V at 203kΩ (12.7µA)
2.31V at 16.9kΩ (137µA)
^ here the computer did not switch into external mic mode

v here the computer did switch into external mic mode
2200mV at 12.1kΩ (0.182mA, ~400µW)
2140mV at ~10kΩ (~0.21mA, ~460µW)
1830mV at ~5kΩ (~0.37mA, ~669µW)
840mV at ~1kΩ (~0.84mA, ~710µW)
^ here the computer started sending current automatically

v here I had to start at a higher resistance and go lower
710mV at 850Ω (0.835mA, 590µW)
510mV at ~500Ω (~1.0mA)
89mV at 81Ω (1.09mA, 89µW)
~40mV at 39Ω (~1.0mA)
The curve shows pretty much a 2.61V source with a 2.37kΩ resistor in series, which is a little odd. The maximum power I could draw from such a system would be a little above 720µW (1.3V and 0.55mA with a 2.37kΩ load), which isn't too great.

Note that I have to have the earphones plugged into the other side to get the current started, but once it had I can remove the earphones. If I short the L, R, and Gnd contacts instead of having my earphones plugged in I get the current too, and this works by shorting either single contact with ground too.

>>1177101
I'd like to be able to detect steel, as I don't imagine many people are going to be carrying around iron knives. Where'd you get a relative permeability of 200 for iron? I see 5000 on https://en.wikipedia.org/wiki/Permeability_(electromagnetism)
with carbon steel's at 100. Taking every other assumption to be accurate, the inductance would change by .5%.

I think I'm going to try my hand at modelling this in ANSYS.

Can you please go into more depth about detecting the increased reactance? Right now, >>1177058
seems like the most intuitive detection scheme to me. Although I think it would be easier to set it up as a band pass filter with as high a Q factor as possible so that .5% inductance change shows up, such as is shown in figure 4 of https://en.wikipedia.org/wiki/RLC_circuit . I'm still looking for the expression for the Q factor of that circuit. A high Q factor would probably be the most important value, so it would drive the inductor value and capacitance as long as I can find active components with enough bandwidth.

how do I get good at electronic circuits? I'm almost graduating in electronic engineering but I don't know shit about circuits.

>>1177260
When is Devry's graduation?

File: 9nnnw4D.gif (46KB, 225x225px) Image search: [Google]

46KB, 225x225px

>>1177293
>>1177293
>>1177293
>>1177293

NEW THREAD

>>1177194
>>1177208
Microphones are usually biased at 5V. Just take a DC meter and you can read it from any mic in port.