/ohm/ - electronics general - neat opamp applications edition

Making a shitty Modified Sine wave inverter for my graduation project.

>>946499
what the fuck is this

>>946509
ECG, faggot

>>946521
thanks

File: DSCN3506[1].jpg (489KB, 1600x1200px) Image search: [Google]

489KB, 1600x1200px

>>946499
>What are you working on, /ohm/?
The "current projects thread" is all woodworking so I'll start putting updates here.

I'm building a "power glove" that makes sound and video by interacting with a CRT monitor through phototransistors. I got the video effect running yesterday: https://www.youtube.com/watch?v=Uvco72Abdss

Presently I'm building a new audio effect board. The speed of a cassette tape is controlled by horizontal finger position (forwards and reverse at various speeds). Audio from the cassette player goes to a ring oscillator (Dalek voice thing) which is controlled by vertical finger position.

>>946499
what the fuck is all the other shit before going to the ADS8321. I understand some amplifiers and active filters, but the way the circuit was laid out makes it fucking unreadable

>>946521
different anon, I thought it was estim at first.. I was like "what an idiot"

File: 1455488655351-2041377105.jpg (1MB, 1944x2592px) Image search: [Google]

1MB, 1944x2592px

What kind of trimpot is this?

>>946581
surface mount?
screwdriver actuated?

File: 1k-ohm-et3314-smd-single-turn-trimpot.jpg (9KB, 250x250px) Image search: [Google]

9KB, 250x250px

>>946581

Could it be a 3314?

>>946581
a 2M ohm one

>>946633
kek
try 5k

>>946638
My dumb ass read it as 205 and not as 502 :(

>>946499
Do people here use FPGAs? I was just thinking about getting one for fun

>>946659
FPGAs are great but have a steep learning curve. You'll learn a fuckton about MCU architecture and digital design flow though.

>>946657
That's ok anon don't beat yourself up about it. There is one upside down one I saw and is a really shit font to use. The clue is the 'y'
I guess they rely on you checking the datasheet for orientation which is stupid

what are some good kits to get started with analog circuits?

I just want to start making shit but don't know how

>>946504
No one cares tripfag.

https://oshpark.com/shared_projects/kmG2vi7N

Just ordered one of these for my girlfriend.

It emulates the stock CD changer (It communicates to the head unit/information display through CANBUS) and adds bluetooth.

Steering wheel buttons even work with it, along with different button presses on the head unit can enable different features.

CANBUS is cool.

need litz wire for a transformer, but as far as i can tell the only way to buy it is from ebay in limited selection. can i just use multiple parallel magnet wire strands or is there something special about the way litz is woven for suppressing proximity effect?

>>946767
Find an old ATX power supply, the transformers inside of those usually have a good amount of litz wire inside.

>>946659
It was my undergrad specialization and now it's my job. I enjoy it, but it's kind of niche if you're looking for a job in it.

I think they're really fun, but I'm a tad bias.

I just finished a rasperry pi powered ip webcam in a birdhouse to watch the birds grow. It also uses the GPIO pins to turn on an IR led at night so I can see in the dark.

>>946778
Do you hate all the comp sci people trying to stuff FPGAs with shitty high level code?

>>946784
That was more of a problem during undergrad. The biggest trap for newbies is it's hard to break the habit of programming things as if it always runs procedurally.

At work what occasionally happens is upper management will need a simple module asap to patch something last minute so they'll have the IT guys write it in c then compile it to system verilog and just recompile our project with that new poorly optimized module without telling us. 9 times out of 10 it doesn't really matter but every once in a while they'll either run up the latches on the chip or slow the system out of spec then we get blamed for it. It's only happened twice, that I know of, in my time here though.

File: arduino-solenoid.jpg (38KB, 1746x908px) Image search: [Google]

38KB, 1746x908px

I got my solenoid capacitor discharge circuit to work as I want it to, so now my next step is to find a way to make it so that the arduino can be powered without a computer USB connection.

Is pic related the right way to connect the arduino to the battery so that the capacitor discharge system and the microcontroller are all running from the same battery?

File: electromagnet-nail.gif (6KB, 213x136px) Image search: [Google]

6KB, 213x136px

I'm trying to figure out how to make an electromagnet powered by a single AA battery.

I need to know the resistance necessary to not blow the battery and...formula for needed copper wire gauge , radius of iron center and coils needed to reach needed resistance while maximizing strength.

Are there any formula like that?

File: smug maxwell.jpg (64KB, 1280x720px) Image search: [Google]

64KB, 1280x720px

>>946810

>>946810
>Are there any formula like that?

the real problem is that there are entire books filled with formulas from beginning to end, and you have no fucking idea which one to use unless you've spent months studying the subject.

so you can either re-invent the math by trial and error, or follow someone else's design.

(stupid Firefox thinks else's is not a valid word)

>>946806
Usually arduinos have an onboard linear regulator, so you can power it straight from your 10v source. Check it for your specific arduino version though.

File: help.png (143KB, 861x721px) Image search: [Google]

143KB, 861x721px

>>946499
Help, can someone explain this to me? This doesn't follow german norms and design rules so it is confusing me, IN is input from the instrument, OUT is the output, no problem here. But what does SW and VB mean? And what the hell are those pin symbols? I'll also get rid of the LED, but that won't change anything. I'm making my own PCB for this.
http://www.madbeanpedals.com/projects/Afterlife/Afterlife2015.pdf

>>946983
SW is probably some kind of switch. The pins are the through holes on the PCB layout. For example, the pin on C4 is pin 3 on the board. VB is another name for the voltage across the noninverting input of the opamp and ground. Similarly, VA would be the inverting voltage. All of the VB connections go to the noninverting input labelled VB. Those terms come from Vo = Avol(Vb-Va)

>>947003
>SW is probably some kind of switch.
That makes a lot of sense, the LED is controlled by the switch on this thing.
>The pins are the through holes on the PCB layout. For example, the pin on C4 is pin 3 on the board.
Oh, I just realized that they did it with a double sided PCB, again, makes a lot of sense now.
VB is another name for the voltage across the noninverting input of the opamp and ground. Similarly, VA would be the inverting voltage. All of the VB connections go to the noninverting input labelled VB. Those terms come from Vo = Avol(Vb-Va)
Oh, interesting. So all the VB lables literally just go into the OP-amp? I assume that's done to make the schematic easier to read/eliminate a bunch of crossing wires?

Thanks a lot! That was really helpful.

>>946862
>so you can either re-invent the math by trial and error, or follow someone else's design.

Alright. Damn this is getting expensive quick.

>>947003
>>946983
Okay, so here is my version of it in eagle. I know it's a lot to ask but if someone has the time and nothing else to do then it would be awesome if they could check it and see if I made any obvious mistakes. I'll obviously do the same but you never know.

Also, where exactly does pin 8 and 4 on the second op-amp go in that schematic? Maybe I'm overlooking them but I can't see them in the schematic and eagle won't let me invoke those pins on the second op-amp either. Oh and I only have one IC on the board now if I generate it from this schematic, which is also weird. Or are there 2 op-amps in one DIL8 package?

>>946499

http://visual6502.org/sim/varm/armgl.html

File: stuff.png (43KB, 3789x1947px) Image search: [Google]

43KB, 3789x1947px

>>947043
I'm stupid, forgot the pic of course. And I figured the op-amp thing out, it's 2 in one package so it only needs those pins once.

>>947050
i don't know anything about audio
and i don't know much about analog
but i do know that tl072 is two opamps in dip8 and tl074 is four in dip14
and that if you drive them to negative rail they will output positive rail
the schematic looks stupid to me but like i said i'm no analog guy. then again i know a lot of audio guys aren't either and lots of circuits are just a heap of shit blindly put together in random combinations until they do something interesting and end up including unecessary bits and bobs here and there. but sometimes they are very clever too. so be aware.

>>946862
Or just assume ideal properties and use ohms law.

>>947055
Well, what I'm doing here is basically blindly cloning the schematic and then just designing the PCB. It's a validated design, several people have built it like this and reported that it works, so I'll just hope that it'll do the same for me honestly. Not brilliant, I know, but I'm not looking to change or improve upon the design right now, I just want to build a working compressor. Thanks again for the help, now the only remaining issue is that the opto-coupler is not available from my electronics retailer, only from a music store, so that means extra shipping costs for me if I can't find a substitute with the same specs.

>>946810
>a single aa battery
its not going to be very strong. what are your plans for it?

>>947082
Just to move another magnet along a rail.

>>946810
Just to make sure, that pic of yours is meant for classroom demonstrations. It is NOT meant to show how to build a strong electromagnet and it is weak in comparison to more sensibly built magnets.
And like anon already said, there are whole books written about the topic. Some filled with theoretical stuff, some with more cookbook like approach.

>>946932
What about things like voltage drops/spikes due to the inductive load and boost convertor?

>>947084
Get some enamal coated wire* and wint it around a nail.
* = iirc enamal coated wire is what works good for magnets, been a long time since I messed with any of this stuff, double check it.

>>947106

Yes

enamel coated wire is used for anything magnetic because otherwise coils would short against themselves and their ferrite cores

>>947087
Yeah, I just posted the most basic Googled image I could find so people interested in electromagnets would be able to find my post easier.

>>947106
Ordering bunches. Now I just need to find a place that can mill iron.

>>947117
> Now I just need to find a place that can mill iron.
Ideally, you want either soft iron (if you're only using DC) or silicon steel (for AC).

Soft iron doesn't become permanently magnetised. Alloying with silicon increases electrical resistance which reduces eddy currents.

File: 2016_02_15_12.41.14.png (178KB, 1080x1920px) Image search: [Google]

178KB, 1080x1920px

>>947043
You should probably look up the datasheet for that op amp. The one in their PCB design is an 8 pin so it has two opamps in that package. The square is pin 1. You could probably figure out all you need to know by following the traces on the PCB design and comparing it to their schematic. The colors tell you which side of the board it's on.

Worst case, you could always throw it on a breadboard and see if it works. Failing is the fastest way to learn.

File: time consuming piece of.jpg (125KB, 1741x979px) Image search: [Google]

125KB, 1741x979px

>>946499
>>946521

> Museum comes asking for a interactive instalation where people put their hand and see their heartbeat in some leds, for great cash
> said sure, i did much more complex digital projects. these idiots have no idea of electronic projects complexity
> start research and find out lots of examples like OP
>"welcome to the anonmuseum! cover your hands with this gel and put them here. Now remove one of your shoes and step in this gel bucket" not gonna happen
>Try making a 2 probe version with active electrodes
>beautiful 60 hz noise no mater what combination of filters
>CNCing fancy inox shielded electrodes
>beautiful 60 hz noise no matter what
>fml
>opening day coming, client start to suspect something wrong
>show them video of leds beating with pwm control, following some mp3 heartbeat from google
>"its wonderful better than expected can we test"
>fml harder
>import expensive biosignal ampops
>beautiful 60 hz noise no mater what
>about to give up and return money
>look at home treadmill no one uses
>hmmm this thing has a hands only heartbeat sensor let me see
>2 inox plate in each hand, wtf
>open up to find cute little 6 pin module soldered in main board, with anon epoxy blob chip
> module has 2 inputs from each hand, 1 return signal shared to other pair of metal plates
>3v3 power pin, gnd
>last pin is 3v square wave signal of heartbeat
>GodExists.jpg
>dessolder from board, wire to microcontroller, it works with any shit electrode, even rusted pipes
>client loves end result, hires me to 3 other installations, doesnt even look inside circuit box

Dont want to mess with biosignals ever again

Pic related

>>947159
>Museum finishes exhibit
>Curator is friends with Dave Jones from eevblog
>Sends project to Dave
>Dave does a teardown
>Bye bye job

File: ohm trolling device.jpg (1MB, 3338x5000px) Image search: [Google]

1MB, 3338x5000px

>>946810
Just do it with a resister in series with it. Decrease the resistance as needed. Better yet use a PWM rated at whatever amps you need.

>>947159
please be true

File: IMG_20160215_154530.jpg (2MB, 4128x3096px) Image search: [Google]

2MB, 4128x3096px

Hi all.

I ran across this keypad, and managed to figure out which paths close when each key is pressed, but I'm not sure how to go about reading this with an arduino or raspberry pi, since I can't see much for a common ground, besides many pins f,e,g,h. But at the same time, that'd be hard to read if they were all marked as ground, because I'd have no way of seeing which ground was used. Confused. Please help?

>>947159
That's what I would have done in the first place. Though, I would have got the parts from sparkfun (just to search it), mouser, and/or digikey.

Don't cringe too hard.
https://youtu.be/fPODzvU3Iwo
https://www.sparkfun.com/products/11574

>>947160
Permanent exhibit more than 10000 km away from murica. Also was while undergrad so they were paying much less than for a senior engineer, still lots of money for poor undergrad me.

I still want to learn how to diy them properly. Any tips? Or where to find these modules in bulk?

Just needed a robust solution fast and i doubt any diy thing would be better than something embedded in a standard product like a treadmill, with several years of development.

File: comp.png (23KB, 651x660px) Image search: [Google]

23KB, 651x660px

>>947129
>>947055
Thanks again, layout is done. Not perfect but it should hopefully do the job. One last question, does anyone know if the capacitors need to be closer to the IC or any other components? IIRC some caps need to be physically close to something to work properly.

>>947160
But would be an great honor to have Dave analyse my undergrad diy stuff. Lame solution but still works after several years and thousands of greasy hands.

>>947172
I did this before, internet is flooded with arduino examples with ir leds and phototransistors. Just coudnt get robust results, always had to adjust fingers a lot.
Fancy bio amp ops came from mouser, are still somewhere around here

>>947176
> does anyone know if the capacitors need to be closer to the IC or any other components? IIRC some caps need to be physically close to something to work properly.
Decoupling capacitors on the power rails of digital ICs need to be close to the IC. But that isn't relevant here.

>>947171
It's an X-Y matrix. Pressing a button closes a contact between one row pin and one column pin. Looking at your table:

E = Row 1 (Top)
F = Row 2
G = Row 3
H = Row 4 (Bottom)

A = Column 1 (Left)
B = Column 2
C = Column 3
D = Column 4 (Right)

To read it with a uC you put a drive signal on a single row or column (A-D or E-G) and read the switch state back on the other 4 pins. You have to do this on all four rows/columns in turn, scanning across the keypad for closed switches.

>>947185
So, the op-amps don't care basically? Good to know, thank you. The cap itself is actually close to the IC but the ground trace is way too long if that was relevant, I think, don't tell anyone that I'm graduating as an EE in a year.

>>947189
Thanks for the help, it was right in front of my face. As far as the method to actually read which key is pressed, you're saying I should put a signal out on all of the pins for the rows, then read for the columns? Then put a signal out on all the colums and read which row gets a signal?

Just want to be sure everything's hunky dory before I go wasting time with code, yeah?

>>947176
if C1 is supposed to be your decoupling cap then thats pretty piss poor, could do better.
if C5 is supposed to be your decoupling cap then thats pretty ok actually

>>947185
well decoupling isn't just for digital ICs, anything that switches can produce transients on the supply which is perfectly possible for an opamp depending on the output signal.

>>947159
You should've used a n instrument amplifier and make the 60hz your common mode signal then amplify the difference. That's how ECGs are done.

>>947201
no!

you drive rows high/low as outputs and read columns as inputs (or vice versa but not both!!)

say you press button 4

you drive only E high then test each of the columns. all low
you drive only F high then test each of the columns.
input B is high
F + B = 4 from the table. boom, done.

lets do another one.

press button right arrow
drive only row E high
test all inputs, all low
drive only row F high
test all inputs, all low
drive only row G high
test all inputs, all low
drive only row h high
input B is high
H + B = right arrow. boom. done.

you do the driving and the testing in a loop and break out when one goes high.
alternatively if you have interrupts you can just set them all high then service an interrupt when the input changes, during the interrupt run the scan to find out which button is being pressed.

also use pulldown (or pullup i guess) resistors on the inputs so they don't float.

>>947206
>f C1 is supposed to
>if C5 is supposed to
I, erm, don't know. The cap parallel to the power delivery of an IC would be the decoupling capacitor, right? In that case that would be the one in the bottom right, directly in the corner. It's close to the IC but the ground plane is way too long. Fuck, back to the drawing board I guess,

>>947211
This.

(Just finished typing...)
Eh, not quite. Lets say the 4 rows A-D go to 4 output pins on the uC, and the 4 columns E-H go to 4 input pins. First output a 1 to a *single* row on the matrix and a 0 to all the others. Now you can read the state of all the keys in that row on your input pins. Then you output a 1 to the next row and so on. Note that normally you'd need an external pulldown resistor on each of the uC inputs for this to work. Your particular uC may have these built in.

>>947159

how do you get that kind of job building custom stuff?

I am amazing at building whizzbang stuff like that

>>947213
> The cap parallel to the power delivery of an IC would be the decoupling capacitor, right?
Yes. A decoupling capacitor is basically a reservoir of energy "local" to the IC, so that brief spikes of current draw can be supplied locally.

Essentially, at high frequencies a PCB trace (or a wire) stops behaving like the theoretical model (where the voltage is the same at all points) and starts behaving like a string of resistors and inductors which get in the way of the current flow, particularly at high frequencies.

From the schematic, it appears that both C1 and C5 are decoupling capacitors. It's not uncommon to have multiple capacitors for different frequency ranges (larger capacitors typically have greater parasitic resistance and inductance which limits their effectiveness at high frequencies). It's more important for the smaller capacitor to be close to the IC.

This stuff is more important at higher frequencies, and more important for switching circuits (i.e. square/pulse waveforms with high dV/dt and dI/dt at the edges) than linear circuits. It matters a lot for complex digital logic (where most of the current draw is on clock edges) dealing with square waves at MHz or even GHz frequencies. A CPU may have dozens of decoupling capacitors, each within a few millimetres of the socket. It matters more for a large, high-density PCB with long, thin traces.

For an audio circuit (particularly with the fat traces of a PCB designed for 0.1" DIP), you probably wouldn't notice any difference if you removed them altogether.

>>947213
well all i can see c1 and c5 doing is decoupling ac noise, c2 as well actually.
if you run it from 9v wall wart then fine, 9v battery you probably dont need c1
the point of it being close to the ic is to not introduce extra inductance in the leads. if you have a nice big juicy ground plane then you dont have to worry so much, i mean its not ideal but it will do. usually if you are serious about decoupling you put an smd chip on the backside or under a dip. its not like you can put a cap through the package so supplies on opposite corners are a pain.

id say its breddy gud.

>>947231
Get out of engineering building and become friends with artists, biologists, anyone who studied things far from what you do. And truly appreciate what they do and learn from them. I had too many class mates with that dick posture of "stem rules, bio and human sucks".

You will see how your technical knowledge inspires them and how their imagination challenges you. Awesome custom stuff comes from this match and there's plenty of people willing to pay for.

File: IMG_20160215_173228.jpg (2MB, 4128x3096px) Image search: [Google]

2MB, 4128x3096px

>>947218
>>947211
Ok, thanks for that help.
I have a "working" model, but it also gives a lot of erronious results, mostly probably because of floating values.

I understand how a pull-up/down resistor is supposed to work, but in this case where we are changing the values quickly, I'm not sure how to implement it in order to get rid of the floating value messing things up. I'll post an image of the breadboard and one of the code I wrote.

Help me along a little more? Thank you.

File: Screenshot from 2016-02-15 17-35-25.png (49KB, 475x595px) Image search: [Google]

49KB, 475x595px

>>947258

File: Polar heart_rate_monitor_sensor_belt.jpg (28KB, 321x600px) Image search: [Google]

28KB, 321x600px

>>947159

like the other dude said, you must use a 3-wire instrumentation amp - the common goes to both hands, and the + and - inputs each goes to one hand. you follow that by a very high gain stage, followed by a comparator. oddly enough, you can do this with crappy LM324s and it'll work just fine.

optical solutions are shit coz you gotta set your finger just right. if you must absolutely have a 2-wire solution, then the chest straps made by Polar will work. about 10 years ago, they used to transmit a magnetic pulse (not radio) to a watch, but i guess these days they've probably gone bluetooth.

pic shows an old unit, with 3V cell at the bottom, and a ferrite core to the left of the PCB.

File: Screenshot from 2016-02-15 17-38-10.png (13KB, 82x995px) Image search: [Google]

13KB, 82x995px

>>947259
Here is some output while holding the "3" key. It should only output the numbers "37 15" while it's pressed.

As you can see, it works half the time, and does not the other half.
Is it my code?

>>947243
>>947249
Alright, thank you so much. I'll either redesign the PCB or add a jumper wire from the pin 4 on the IC directly to the capacitor on the groundplane, essentially skipping the long trace around the entire thing. But wouldn't those caps usually be ceramic instead of film? Or is that only for high frequency? Because everyone I've seen built them with film caps. I assume they can't be that important, the highest (fundamental) frequencies I'll be dealing with are 400Hz on bass + harmonics.

File: sweet_cable_iuyiouoyiuo.jpg (19KB, 270x379px) Image search: [Google]

19KB, 270x379px

while roaming aliexpress I found a few places selling cable like this, which kinda makes my benis hard

I haven't *ever* seen cable like this in use tho.... what is it used for? anybody know???

Most of it I found is rather thick, like 3/4" / 20mm or greater. The cost doesn't seem particularly high, but all the places selling it have huge minimum orders. :|
Usually there is one central conductor area, with groups of others arranged around it. I found sizes up to around 2" / 50mm diameter

>>947260
Thanks for the help, the project is from 2012, first thing i tried were instrumentation amps, just like the textbook, but not with grounds in hands. I will try that whenever I feel like going bio again. Anyway back then i just got lots of 60 hz noise. I guess the arms make a loop too big and inside a building with mains everywhere even the difference between 60 hz on right and left was still bigger than ecg signal.

>>947266
film caps can be more linear in frequency response
if you can identify the difference in audio output i'll buy you a cookie

>>947273
>Most of it I found is rather thick
thats the point.
i've never seen 5 core arranged like that but 3 and 4 core cables i see all the time arranged like this
the reason its done is to get as much cross sectional area into the circular cable profile, to maximise the packing density. thats why its thick cable because as it gets thicker the wasted space becomes an issue.

also skin effect for ac is better but in your picture they would be mismatched which is bad.

>>947262
code looks ok, did you use pulldown resistors?

>>947279
>film caps can be more linear in frequency response
But ceramic caps work better at very high frequencies, right? That's how I remember it at least.
>if you can identify the difference in audio output i'll buy you a cookie
I guess that means I don't have to worry, I'll just add the jumper wire to make sure and call it a day then.

>>947258
First thing: the output for the "active" row should be pulled either high or low, while those for the inactive rows should be left open-circuit (i.e. either configured as an input or driven through an open-collector buffer or transistor). The pull-up/down resistor should pull in the opposite direction to which the active row is driven (i.e. if the active row is driven high, use a pull-down, if it's driven low, use a pull-up).

Second: don't toggle the row output for each column; i.e. move the GPIO.output() out of the "for col_pin" loop. You might also want to add some delay between changing the output row and reading the input columns (although just using Python may be enough of a delay).

Third: debounce. When a switch opens or closes, there's a borderline point where it's barely making contact. In this state, the input may alternate between 0 and 1. And you probably don't want this to be treated as a dozen distinct key presses, so you need to track the history of the key state and only change its state when the value is consistently 0 or 1 for a given period of time (like, more than a microsecond).

File: Screenshot from 2016-02-15 20-36-45.png (62KB, 603x791px) Image search: [Google]

62KB, 603x791px

>>947298
I was sort of able to get where I wanted with this using LEDs as indicators for row/column activity, but I'm still unable to get the damn program to output the correct pair of pins. It's only outputing the correct row.

Here's a video of me with the LEDs:
http://pappad.info/owncloud/index.php/s/R2IMVLStaHUmbWn

File: FPIEXTII67S87QB.LARGE.jpg (82KB, 1024x680px) Image search: [Google]

82KB, 1024x680px

Hey guys.

Making a piezo/arduino drum kit and need to lengthen the wires coming out of the piezos.

The wires on there are thin as fuck. If I want to extend them to a meter or so in length will wire thickness have an effect on anything?

I know jack shit about electronics. I just don't want to start running into issues with lag or anything.

Can I just use 'regular' gauge wire or should I try and find really thin wire?

Thanks

>>947375
I'm this guy.
I figured out the problem I was having. I hate duplicating blocks of the same code, to me it makes the whole dynamic language thing of python be meaningless. Nevertheless, I just swapped the col/row variables in another read attempt block.

Python is fast enough to read both values of a momentary touch.

Thanks for the help today guys.

>>947375
what god forsaken language is that, python?

>>947404
What's wrong?

>>947172
what the fucking fuck, is that thing on the left a guy or girl?
if guy is it on hrt or what

>>947405
nothing in particular, just inquiring about the language in that anons picture.

>>947406
A girl with a glue-on beard.

>>947409
ok, all is right again
I thought the future was finally here and was about to freak out

>>947411
nah, hrt can't reverse the effects of long term aging with your natural hormone balance. That would be like a decade of gene therapy which is many decades off and will probably be illegal anyways. Feel comfortable.

>>947413
The one on the left was taller. the way these scrawny ass young "guys" look and act now, I expect will be considered a "man" within 100 years.

>>947417
Yes, we just need more xeno/phyto estrogens daily.

They've been lowering the "normal" range of male testosterone levels for decades. In the 50s 700 was average, now 200-700 is normal. Mostly people fall in the range of 450. The documentary The Disappearing Male is good material on this subject.

>>947420
>The Disappearing Male
I knew the normal range was in the hundreads, I didnt know they were lowering the standard though.
Im going to look into that documentary.
There need to be more people aware that our gender is under attack.

File: Untitled.png (4KB, 415x587px) Image search: [Google]

4KB, 415x587px

Trying to learn electronics on my own and I have a problem understanding capacitors in series.

Capacitance is the amount of electrons stored on the negative plate right? So with parallel capacitors it easy to understand that total capacitance is C=C1+C2 etc. But I don't really understand why capacitors in series store charge of C=1/(1/c1+1/c2+1/c3). Wouldn't the negative plate of C1 first get full charge, let's say 10mF, and C2 would be empty? But according to the equation the total capacitance would be 5mF.

Reading getting started in electronics currently btw.

>>947275
It is normal to have some 60Hz in the signal, but if shit's totally buried in it, then something was wrong.
Anyway, you can filter ECG heavily without affecting the shape that much. For example, 50Hz (or even lower) steep low pass filter can be used. Or you can use a notch filter to kill 60Hz and its harmonics specifically.
Also, if it's not a floating device, isolating the ECG parts with optoisolators or whatever makes things much easier.

>>947260
>optical solutions are shit coz you gotta set your finger just right.
Well, you'll have massive baseline wander whenever you move your finger, but it isn't particularly difficult to design them to be reasonably tolerant to finger positioning. If it's just for heart rate detection, high-pass filtering can reduce the effects of slow movements quite a lot.

File: 09870978.png (19KB, 434x393px) Image search: [Google]

19KB, 434x393px

>why are my IR leds so dim on camera?

Do I have the wrong type? Did I mess up the schematics?

It's for trackIR, a head tracking thing that follows three ir led dots and determines the position of your head. I've thought about making some diffusers for the bulbs but first the actual leds aren't as bright as they should be?

>>947423
>But I don't really understand why capacitors in series store charge of C=1/(1/c1+1/c2+1/c3).

Because what you're basically doing is putting more space between the plates. Putting two (identical) capacitors in series is equivalent to doubling the amount of dielectric in just one of the capacitors. You gain a higher maximum allowable voltage across the plates at the expense of fewer coulombs stored per volt.

You can't have capacitors in series and only charge one of them, because current must flow through all of them.

>>947438

Check that your diodes actually have a forward voltage of 1.5V. If they're even 100mV higher than that, they'll be much dimmer off a 5V source.

>>947438

20mA is not enough to get a very bright light. your LEDs would be adequate for remote controls, because they will amplify the signal a lot, but probably not for an IR webcam. so, you could
- see if you can get more powerful ones. i have a few dead security cams, so i'd use those.
- double them up
- feed them high-current pulses. i.e. if you pass 40mA for 50% of the time, they should survive since they can cool down during the off time. you can probably go 400mA, 5% of the time.

another thing is your gadget supposedly uses reflective strips to reflect the light at the camera; the LEDs point towards you. presumably this makes the dots larger so the camera sees them better.

File: Screen Shot 2016-02-16 at 06.04.40.png (28KB, 471x536px) Image search: [Google]

28KB, 471x536px

>>946499
has anyone ever done something like this? I'm considering it as an alternative to using a raspberry pi + camera + wifi module since a cheap android phone has all these things (plus a battery) built into one compact package and costs about the same as getting everything separately. The android just needs to be able to run opencv and host a server (node ideally, but I can figure a way to make whatever work). Also I want the android app to read from the camera and run the server while the screen is off (to preserve battery).

Are there any holes in my plan you guys are aware of? Any reasons why the RPi setup would be better?

>>946778
lol that's a good trip

Holy fuck /Ohm/ why do companies make shit so hard to fix?

I've recently bought a tascam cassette recorder and the Capston belt has turned to goo from age. Thought this would be a fairly easy fix so bought the new belt along with the drive belt too.
Literally had to take the whole fucking thing apart Piece by piece to get to the belts, it's like the whole thing was built from the ground up around those two belts!
Needless to say I've ended up with a few problems putting it all back together again.

>>946810
What are ohm's law and resistivity.

Or just do this >>947164

>>947476
Just do it! And post results somewhere.

You are treading uncharted waters, don't ask for more advice here, it seems no one has any to give.

>>947478
They don't want you to repair your own shit, the /diy/-crowd is the biggest nightmare of many companies. Also, sometimes it makes certain parts of manufacturing easier and maybe saves 1cent over making it more accessible.

>>947478
was it common/high end? maybe there's a service manual

>>947496
Yeah think that must be the case here. I guess that's maybe why you don't see a lot of them about. This is quite a tidy looking unit overall so think it'd be a shame to not get it up and running.
What I thought might be a relatively easy fix turned out to be way over my head.

>>947500
Yeah it's a Tascam 112. I found the service manual for it, I just wasn't expecting it to be so much hassle to replace a couple of rubber drive bands. Even after unplugging all the plugs from the main pcb the remaining soldered wires to the drive unit hardly left any room to get access.
Anyway I've replaced the drive and transport bands and managed to pull some wires from the mother off (due to being so tight) gonna try and resolder them back on again later.

>>947423
> Capacitance is the amount of electrons stored on the negative plate right?
No. Capacitance is the change in charge (Coulombs) per volt. Q=C*V, C=Q/V, V=Q/C.

> But I don't really understand why capacitors in series store charge of C=1/(1/c1+1/c2+1/c3). Wouldn't the negative plate of C1 first get full charge, let's say 10mF, and C2 would be empty? But according to the equation the total capacitance would be 5mF.
With capacitors in series, the change in charge applies to all capacitors equally: whatever charge goes in to the +ve plate of C1 also moves from the -ve plate of C1 to the +ve plate of C2, from the -ve plate of C2 to the +ve plate of C3, and out from the -ve plate of C3.

The total voltage is the sum of the voltages: V=Q/C1+Q/C2+Q/C3 = Q*(1/C1+1/C2+1/C3), so the total capacitance Q/V = Q/(Q*(1/C1+1/C2+1/C3)) = 1/C1+1/C2+1/C3.

File: 1404951903144 _g_ android.png (759KB, 1080x1920px) Image search: [Google]

759KB, 1080x1920px

>>947476
i've thought about using an old/cheap android phone for a project (weather station or security camera) but i didn't get to it yet.
i think it's a good idea. if you do it keep posting your progress/ideas

>>946499
You can reduce the complexity of the circuit significantly if you use one of Plessey's ECG sensors (http://www.plesseysemiconductors.com/epic-plessey-semiconductors.php), it's a high impedence electric potential sensor that can be coupled to the skin without any preparation and only requires some filtering and a differential op-amp to achieve pretty good results. Funnily enough I'm working on an ECG at the moment.

File: FT8PPQII0LDVS1P.LARGE.jpg (179KB, 979x1024px) Image search: [Google]

179KB, 979x1024px

I'm building this automatic plant watering circuit. It works in the guy's videos, but I am not sure how. I know the letters are switched on the NPN SS9014 transistor. But the circuit is supposed to turn off when the two soil sensor probes connect (the probes are in the soil and when there is enough water between them the water completes the circuit and the pump turns off). Lets assume that there is enough water so the two water sensor probes are basically connected. Wouldn't having the soil sensor probes connected make the pump turn on, while having them not connected make the pump turn off? It seems like this is backwards can someone please help me understand this?

is it possible to do i2c over uart?

>>947429
Thanks for the tips. I really spent a long time trying to tune the filters in that pic, and it worked with chest wet electrodes, just not dry hand ones. Also in this projects good emc board design is a must and i never had to learn about that before, digital circuits are much more tolerant. Do you have tips in this area?

Also in instalation design phase they asked if could be something that way, putting hands in metal, and i naively said sure.

>>947645
after that museum adventure i really admire you

>>946499
/ohm/ does nichrome wire have to be apart from itself to keep it from shorting out? I'm trying to make a Rope cutter for myself.. but im so confused. can I just squish the wires between a thin piece of plate steel shaped like a V and it will still get hot? or do I need to insulate the wire somehow to keep it from shorting off the plate and itself?

if so, what do you use to keep it from shorting off itself? but still can conduct heat [250degree F.]

>>947755
>/ohm/ does nichrome wire have to be apart from itself to keep it from shorting out?

Yes.

>do I need to insulate the wire somehow to keep it from shorting off the plate and itself

Yes.

If you're getting it red hot, there isn't much that's practical that you could use. The only thing I can think of (which I used for a small shitty oven) is mica.

But that conducts heat poorly and is very fragile. I don't see why you couldn't just stick two bolts through a piece of wood and tie one end to each...?

>>947759
250degrees isn't red hot, hot enough to burn your fingers. but its only a few degrees above water boiling.. for cutting large nylon ropes [towing rope to be exact] also the bolt thing doesn't work out [bare wire] because it gets gunk on the wire itself and is a PITA to clean off. also when cutting through the thick ropes its not getting the center hot enough to melt the inner fibers. So I thought if I could have more surface area heated. it would allow for the fibers in the middle of the rope to melt properly.

>>947661

You could try with a truth table for the circuit, but the use of the soil sensors seem a bit redundant.

>>947737
If you're referring to the OP pic, then those filters are pretty minimal.
Sure, using dry stainless electrodes instead of the usual Ag/AgCl gel electrodes will make things more difficult and using just two electrodes makes things even more difficult.
EMC is a broad topic, but besides the usual stuff I don't really have that many "tips" for ECG. Letting the amplifier part float helps, like I already said. Using a third (reference/leg) electrode helps. And if you try to manage with just two electrodes, carefully balanced and very high impedance inputs help. Active electrodes (buffer amplifiers as close to skin as possible) help. Not hurting yourself intentionally by placing noisy stuff like fluorescent lights close is a good idea, too.

Well, you got it working after the initial frustration. If nothing else, you have now much better idea what is easy and what is not if someone wants another ECG thingy.

>>947704
You could implement bit-bang I2C using modem control lines. If you insist on using just rx/tx, then you need some kind of bridge. I think such bridges are available ready-made, but it isn't that difficult to program a microcontroller to do that either.

File: 300-xl[1].jpg (361KB, 1000x754px) Image search: [Google]

361KB, 1000x754px

What was wrong with the Arduino Leonardo that made the Arduino.cc people discontinue it?

>>947159

>any shit electrode, even rusted pipes

this made me lose hard for some reason

File: 51eeb8f9ce395f0778000000[1].png (293KB, 560x416px) Image search: [Google]

293KB, 560x416px

Hey guys if I want to have more space on my arduino
>pic related

would adding sd card module and having it act as a hdd work or how would I go about that?

>>948404
do you mean you need more space for the flash that holds the code, or are you looking to add storage for data logging or something like that? the latter is doable (here's an example http://www.instructables.com/id/Arduino-DIY-SD-Card-Logging-Shield/), the former would be a lot more complicated but possible if you can execute instructions from RAM, though at that point, with the level of complexity it would add, you'd be better off just buying a better MCU.

>>948409
the pro mini has 16m flash, that should be plenty for code, it would mostly be used for data logging.

>>947454
>>947459
Forgot to reply, I remembered I used to have a toy that had IR leds meant for illumination instead of sending signals, found it at my parents place and borrowed a few. Much much brighter, works fine now.

>>946499
is arduino good or should i hate it like the rest of /diy/

i want to make stuff with sensors and servos/motors etc

>>948434
It's good, you just won't learn as much as you would trying to do things without it.

>>948435
i wont learn much because of the pre-soldered pre-written libraries? that is a plus right now for me

>>947159
Good story

File: hDQaYiXgCke3M1pR.jpg (437KB, 1600x1200px) Image search: [Google]

437KB, 1600x1200px

This is the fan header of a PS4.

Is it an off-the-shelf header I can find converters/splitters for or is it a custom job?
I know its not a header found on normal consumer boards (mobos, GPUs, laptops, checked them all) but I thought maybe it was used for industrial or embedded electronics.

I've searched for ages and can't find anything on it.

File: IMG_0655-min.jpg (763KB, 2448x3264px) Image search: [Google]

763KB, 2448x3264px

How can I solder this? Solid copper wire to a steel fuse array? Im at 400 degrees and I keep heating the metal for a while but everything breaks off.

>>948673
tin first before 'soldering'
if your iron isn't powerful enough, use a mapp torch

>>947661

>DC soil conductivity sensor

My man let me save you some time. You're better off passing AC through the electrodes. Use the electrodes as a a resistor in a voltage divider network and then use envelope detection on the output of the voltage divider.

DC conductivity sensors have a problem with causing corrosion over time so you'll be regularly replacing electrodes. This is fine if you're okay with it but it's so simple to produce a sine wave that there is no reason not to do this.

>>947066
how are you going to calculate the strengh of a magnet with ohm's law dumbass ? now shut the fuck up

>>948673
why not crimp spade connectors to the wires?

File: ss+(2016-02-18+at+09.47.08).jpg (29KB, 597x298px) Image search: [Google]

29KB, 597x298px

How do you wire these with an adapter? I soldered a circuit like with 2 but was wondering if the circuit was incomplete. Would 1 complete it?

>>946810
a double aa has enough internal resistance to be shorted and not "blow up"
9v however do and go off like a .22

>>947380
thin wires are used at the pizeo solder joint so you dont rip them off eaisly. do not use solid core wire- same reason.
solder wire on the ends of the existing wires
you can solder directly to the piezo but be quick about it as to not damage the piezo

>>948673
skuff up the steel w/ sandpaper and apply lots of flux.
try pre tinning the steel separately. if doesnt take then soldering is out of the question.
also make a better mechanical connection. crimp the copper around the steel.

>>948692
chill bro
maybe not only ohms law, but current and #of turns will give strength.

File: trimmer.jpg (97KB, 411x533px) Image search: [Google]

97KB, 411x533px

>Sorry for crossposting from >>948883 but i need to order something ASAP, so I'll repost here (really - I normally wouldn't spam this slow board otherwise)

I'm quite stumped looking for a precise trimmer pot around the lower ohm range (~5-20).

Thing is it has to withstand around 500mA @ 1.25V (~600mW) so I've been looking for >0.75W ones - found one from vishay but now the datasheet got me confused (see pic related). Specifically their mentioning of "maximum working voltage" and "maximum wiper current". Do I have to meet both of them or can i exceed one if i back-off on the other one and vice versa? I'm thinking my 600mW are lower than the rated 750mW so I should be fine but wanted to check with you guys just to be safe. Thanks in advance.

>>948774
Is that an LED strip? If so, 2 will work as each LED is in parallel with the rest.

>>948889
>Do I have to meet both of them
Yes. You can think it this way: the resistance element, as a whole, can take 750mW (274mA) at some specified conditions. If you don't use the whole element, you can't use it at its maximum power.

>>948894
Shit. Makes sense, thx bro.

File: power selector.png (13KB, 511x395px) Image search: [Google]

13KB, 511x395px

>>948142
32u4 is more expensive? Who knows
I'm drawing a circuit with a PIC16F628a to switch 4 relays on and off. I've included a header where I'm going to connect a usb-serial adapter based on the FT232RL. I don't want the relay coils to receive any power unless the board is connected to a 9-12V wall wart, so I figured out I could use an OP-AMP to switch a P-channel mosfet on and off.

It seems to work on that falstad circuit simulator. What do you think, /ohm/ ?

>>948904
Why not just use a PIC18 with built-in USB (e.g. 18F2550)?

Also: why power the relays from the USB? If you power them from the wall-wart, that automatically ensures they won't be powered when the wall-wart isn't connected.

>>948913
>why power the relays from the USB

I don't want to. I want the power to the relay coils to be cut off when the board is powered by the USB port alone. Also I think one 358 and a salvaged p-channel mosfet are going to be cheaper than 4 optos

Also I don't have any 18F2550 in my drawers

>>948930
So what's wrong with his suggestion of powering the relays from the wall wart? What are these optos you're talking about? Why not just control the relays with a bunch of normal transistors?

What comes to that circuit of yours, the voltage dividers aren't dividing anything. You need to be careful how to power the opamp and in general you could replace it with a transistor, couple of resistors and a zener diode.

>>948935

Is my phrasing that bad?

I'll power the relay coils with a P-channel mosfet, but I'll switch them individually with NPN transistors.

I am going to power the board itself with a wall wart, but I also want the MCU to be powered by the USB port for debugging/testing purposes. What I don't want to do is to provide power to the relays from the USB port.

What should I use to design circuits on Android?

>>948893
i tested to see if any electricity was flowing through the circuit. only figure 1 worked but had only .5 watt . These are supposed to run at 4.8 watts. Any idea how to properly solder wire to this. I tried using acid soldering flux with a 40 watt iron. The solder holds onto the copper terminals but perhaps im missing something..

i'm looking at a power factor correction circuit that has two diodes to handle inrush current to the bulk capacitor.

if the circuit has a current limiting thermistor do i really need those inrush diodes? i don't care to do the math on something that's only a curiosity.

File: 28K077__93928.1439836298.1280.1280.jpg (87KB, 900x628px) Image search: [Google]

87KB, 900x628px

>>946499

What kind of AC do i need to drive a transformer?

I'm thinking of just feeding an H-bridge configuration inverter driven with a square wave into my primary coil.

>>949222
square wave AC can work just fine with only a small (10%ish) efficiency loss over sine waves. that's true at high frequencies at least.

File: IMG_0659-min.jpg (704KB, 2448x3264px) Image search: [Google]

704KB, 2448x3264px

How the f%&$ do I wire this to control an AC current? All tutorials are DC

I think from what I saw I feed an incoming hotwire into the further silver pin, than the output connecting to the crest of the circuit coming from the center?

Then the gold is connected with the ground.

This hotwire is connecting to a wall jack box which will be controlled with relays & arduino.

>>949252
is that an LED switch or a bipolar one? i'm assuming LED in which case you can't put mains through it.

>>949254
LED, okay thanks.
any recommendations? Thinking it should be 15A 120VAC (only need about max 8)

>>949221
> i'm looking at a power factor correction circuit that has two diodes to handle inrush current to the bulk capacitor.
Er, what?

it might help if you post the circuit.

File: 1434046090069.png (33KB, 500x249px) Image search: [Google]

33KB, 500x249px

>>949276
i'm looking at a neat bridgeless PFC circuit specifically. Dc and De exist to provide startup current to Co so that the fast recovery diodes D1 and D2 aren't fried. I'm wondering if an inrush limiting thermistor could reasonably do their job by just limiting the startup current.

>>949222
> I'm thinking of just feeding an H-bridge configuration inverter driven with a square wave into my primary coil.
That's how most high-power switch-mode power supplies work ("full bridge" topology; lower-power supplies tend to use topologies with fewer transistors).

>>949278
I see. The answer is "maybe". The main issue with relying upon limiting inrush current is that the limit may need to be quite low, meaning that a) the circuit takes a long time to start up, and b) it becomes a factor in normal operation, causing the input voltage to sag and the thermistor to dissipate non-negligible power continuously.

Da, Db, Dc, De form a bridge rectifier. They will ensure that Co charges to at least the rectified peak supply voltage without Q1 or Q2 ever being triggered. Once Co is above the peak supply voltage, Da through De will never conduct; they're only used for startup.

L1,Q1,D1 and L2,Q2,D2 are boost converters which will boost Co above the peak supply voltage. A boost converter relies upon the output voltage being at least as high as the input voltage, otherwise the coil and diode will conduct continuously until it is.

D1 and D2 need to operate at high frequencies but only need to be sized for the continuous load current. Da through De operate at mains frequency and need to be sized for the inrush current. The larger Co, the greater the magnitude and/or duration of the inrush current (but operating current is unaffected).

The fact that the inrush current is transient means that it only has to be below their peak current and power dissipation, it can exceed the RMS/mean values provided that the duration is short.

To answer you're original question, you'd really need to simulate the circuit with an accurate model of the thermistor and measure the power dissipation of D1 and D2.

>>949280
That usually works fine if the transformer is designed for an SMPS. A normal AC line transformer, especially if it is a cheap one will probably heat up and maybe make some noise. Should be fine for low power though.

File: relay.png (370KB, 1473x749px) Image search: [Google]

370KB, 1473x749px

Can someone help a noob? I want to control this relay with an arduino, but I can't figure out how to trigger the switch

I think if you run 5V dc through pin 1 and 8 it should flip the switch, allowing AC to go from pin 3 to pin 4, but no success

>>948681
>>948799
Good advice but didn't work

>>948765
Caved and spent the 2$ on connectors,easy but not as satisfying

>>949378
> I think if you run 5V dc through pin 1 and 8 it should flip the switch, allowing AC to go from pin 3 to pin 4, but no success

1. There are different models with different coil voltages (5, 6, 12, or 24V). You need the 5V one.
2. Pin 8 is positive, pin 1 negative. Coils don't have a polarity, so this indicates that it includes a flyback diode in parallel with the coil. If you connect it the wrong way, it will effectively be a short circuit (i.e. it won't activate the coil and may damage the IC or transistor driving it).
3. The 5V model has a coil current of 60mA, which is probably more than the arduino can source or sink directly, so you need to drive it via a transistor or a buffer IC.
4. The 2114 has both normally-open (NO) contacts (pins 3-4) and normally-closed (NC) contacts (pins 6-5). The 2014 has two NC contacts while the 2214 has two NO contacts.

>>949387
>Coils don't have a polarity, so this indicates that it includes a flyback diode in parallel with the coil.
More often than not, there's no diode and the polarization is done using a magnet. In other words, better not omit the external diode.

Bought a soldering iron from Lidl
Swappable tip, tempreture control

>>949413

congratulations, you are well on your way to being the next Wozniak.

>>949413
diy an oscilloscope. it's a nice easy starter project.

>>949539
>>949413
I'd actually recommend starting with a power supply. It's easier and you can't really do much without power.

>>949566
this, a 5-10kw smps is a good starting point.

>>949514
>>949539
>>949567
Off yourselves

Anyone ever manage to get free samples from a manufacturer with a fake company name (or by being honest and just saying you're a hobbyist)?

>>949701
Yes, some companies (quite many) aren't that picky.
Things have been getting worse and worse, so don't ruin things for everyone by pointless hoarding.

What is /diy/'s opinion of adafruit?

>>949713
I'm still pretty beginner but seems alright.

I think it's good for getting newer people interested and helping them make something cool, but you're definitely not 'hacking' though

Prices seem high but not as high as I was expecting.

>>949713
Adafruit is the Buzzfeed of electronics.

File: Adalight.jpg (1MB, 1050x1279px) Image search: [Google]

1MB, 1050x1279px

>>949713
Make their things if you like them but don't pay their prices, just buy the components elsewhere. I use adalight on my TV and I fucking love it.

Replaced the rubber bands with zip ties recently

Anyone converted the adalight code to work with processing 3 yet?

Can someone recommend me a relatively cheap robot kit? There's some $50 ones on ebay and the Zumo line from Polulu look decent. Anyone try them?

File: selsey1.jpg (42KB, 404x304px) Image search: [Google]

42KB, 404x304px

What do you call oscillators (in the AM radio range) which can have their oscillation frequency adjusted?

>>949960

oscillators whose frequency is variable are called oscillators, in the same ways shoes which come in different sizes are called shoes.

maybe you're thinking about voltage controlled oscillators (VCOs), or tune-able local oscillators, like they have in radio reception.

>>949978
>oscillators whose frequency is variable are called oscillators

Nope. Oscillators tend to be fixed on some frequency. I'm looking for an oscillator that can have its oscillation frequency changed.
For example, I have a 1 MHz oscillator right now. It's all good and works well, but I want an oscillator which can have its frequency changed (that is, that isn't just fixed on 1 MHz, but can have the MHz value changed).

>>949986
The generic term in radio use would be VFO, variable frequency oscillator, but it sounds like you want something more specific.
You can implement a VFO as a Colpitts, Hartley, Meissner, etc. oscillator by replacing the capacitor or inductor with a variable component. If this is done electronically, you have'll a VCO, voltage controller oscillator. There are many other options as well.

>>949994
Thanks for the feedback.
What's the traditional solution for handheld AM radio transmitters with adjustable transmission frequencies?
I have a simple AM transmitter soldered, but it uses a crystal oscillator, so it's only fixed to one frequency. What could I replace it with so that I could change the transmission frequency? Some small component would be nice.

File: AM Transmitter.jpg (2MB, 1089x1846px) Image search: [Google]

2MB, 1089x1846px

>>949996
changing out the cystal osc with a variable osc won't work in that circuit. they do a "trick" to get amplitude modulation. they change the voltage on the power connection on the osc module. you won't be able to do that with a different osc. you need to find another am transmitter circuit like this one.

>>946504
>graduation
>shitty
>swift
I hope you fail

>>946543
all those components doing something shitty...

Let's say you have a small amateur pocket radio transmitter. In theory, can you dramatically extends its range just by using a big antenna with a power amplifier?

>>950011
yup

>>950001
Now that's a rude thing to say.

>>950030
don't worry senpai in my experience your senior project can be a burning trash can with a blinking LED swastika on it and you'll still graduate

Do radio signal amplifiers that are used for transmitters amplify the current or the voltage in the antenna? Or do they amplify both?

>>950077
both

File: Doorbell.jpg (103KB, 1142x664px) Image search: [Google]

103KB, 1142x664px

I fixed my doorbell today!

Have literally no experience with electronics, basically just what I had in school.

This is an extremely simple circuit when drawn as a proper diagram, but I was facing two cables that had two wires each, a connector box and a disconnected inductor, so I sweated over this for two and a half hours.

That's pitiful, I know. But I willed it into function.

So I went from no doorbell because the previous tenant disconnected it or something, to having it working again from sheer force of will and what I learned in school.

Feels good, man. :)

Hey /ohm/ does anyone have a good guide to DIY USB hubs and Ethernet switches? I need to put some computer ports in odd places.

The Ethernet needs 1 jack in the usual place to plug into the router, 1 internal connection to the mobo, and 4 empty ports mounted on perfboard along an unused video card slot.

The USB needs 2 hubs mounted in the tape deck and in the back. The back hub should accept PS/2 mouse input and convert the signal on the board.

>>950030
It's okay Swift, I still like you.

>>950270
>namefag sucking off a tripfag
so fucking what

>>950271
I get where you're coming from, but in generals like these I think it's OK to have names if it helps people understand your specialization or you actively post projects.

Of course, off topic posts like this aren't a good use of it so you got me there.

>>950273
if you want people to understand your specialization then make a profile on fucking stackexchange

if you want people to follow you projects buy an abandoned silo from your government and show us pictures of renovating it. or something else equally great. but it doesn't require a trip or name because you will be recognized if necessary from your project.

File: 2000px-Parallel_LC_Circuit.svg.png (39KB, 2000x2000px) Image search: [Google]

39KB, 2000x2000px

Can an LC circuit be used to generate carrier waves for an AM transmitter?
Also, how to make the oscillation frequency of a LC circuit adjustable? Will installing a variable capacitor instead of the standard one do?

>>950307
> Can an LC circuit be used to generate carrier waves for an AM transmitter?
It can be used as part of an oscillator.

You also need an amplifier and some form of automatic gain control (the feedback loop needs a gain of exactly 1).

> Also, how to make the oscillation frequency of a LC circuit adjustable? Will installing a variable capacitor instead of the standard one do?
Variable capacitor or variable inductor.

File: SeriesRC.gif (657B, 261x139px) Image search: [Google]

657B, 261x139px

For an RC charging circuit, is it necessary for there to be a resistor?

As in would the internal resistance of the battery, small resistance of the wires, and the internal resistance of the capacitor be enough?

>>949380
some metal don't work with common solder, complain to mendeleev

>>950249
congrats

>>950336
Your capacitor will charge within a few nanoseconds without the resistor. Is that your goal?

You'd need to use a very large capacitor (Farads) for this to cause the battery to overheat from too much current draw.

>>950345
Yes fast charge is good. I'm using a 25 ohm resistor right now and the power source is actually a 60V DC boost circuit.

So I take it the resistor is there as a safety measure to prevent too much current draw?

>>950348
Also, it looks like I'm at the limit for the current draw of the boost circuit if I use a 25 ohm resistor.

Anyone have good plans for a homemade one? The eBay ones all come from china where everything has shut down for their spring festival.

>>950342
Thanks a lot.

I scratched my head when my multimeter refused to volunteer which of the wires lead to the physical switch outside my door. I had to disconnect it, hotwire those wires together and then try again to realize there was a wiring issue there too. I love the beep function, it's great for checking which circuits close where.

Imagine the look of confusion on my face as I realized there were 12V of AC running through one pair of the wires. At that point I checked the breaker box and discovered the truth: I had no idea how doorbells worked before that point.

>putting on latex gloves before touching any of the wires from fear it was 240V 16A
>mfw 12V 1A

Dunno if normal latex gloves help insulate me sufficiently.

>>950336
You need to limit the current somehow.

If the amount of power involved was significant, you'd use a buck converter rather than a resistor. But a resistor is far simpler and cheaper.

>>946499
Hey can someone redpill me on spark gap igniters? Is it possible to have a constant spark for a while from a regular small battery, maybe a 9V?

I have a few project ideas where one could be useful, but I know nothing about them, and my Google Fu has failed me.

>>950430
The main problem with a constant arc is the amount of heat generated tends to vaporize the electrodes fairly quickly.

>>950449
All right. Thanks for the info. I'll revise that design, maybe just add it to the trigger.

Can someone tell me the basics of how they work and how to make one?

File: Hand Held Butane Gas Pilot Igniter.jpg (12KB, 360x360px) Image search: [Google]

12KB, 360x360px

>>950465

google has plenty of schematics, tho most are powered from 120V. one simple way to generate sparks is with a dollar-store igniter. you dont need the butane, just the electrics.

another option is to use the 4kV output from a disposable camera flash. the gap needs to be kinda small tho, maybe .05 inches.

>>950521
Dollar store igniter uses piezo crystal you moron!

File: product_Sixnet_0605_medium.jpg (5KB, 150x107px) Image search: [Google]

5KB, 150x107px

>>950264
For the Ethernet it looks like I can use something like pic related and connect the power and ports to a perfboard.

The USB seems much easier to do by hand according to

http://www.instructables.com/id/Four-port-USB-hub/?ALLSTEPS

and I just need a USB 3.0 capabe microcontroller from TI.

Still looking for some nice guides to this stuff.

>>950534

and what difference does that make? the guy wants a spark, and this provides what he wants. and you never need to change the battery.

>>946499
Anyone have any experience on the HB 100 ??

>>950547

in both cases, all you have to do is take an existing gadget, pull off some of the connectors, and add extending wires. for the PS2 jack, there are ready-made adapters to USB. there's no need for guides, since you're just changing the housing, like Ahmed did with his clock.

>>950552
Fair enough, I looked at some embedded ethernet chips and almost shit my pants in confusion. Way easier to rip the case off a cheap switch and move stuff around.

TI has a good 4-port USB 3.0 hub reference design that I'd like to attempt.

http://www.ti.com/tool/TIDA-00288

But it's still easier to start backward from an assembled gadget.

My name happens to be Prince Ahmed al-Uruk Allah al-Farook bin Laden, so it's good I'm not in school.

How can I make an inverter for driving this?

http://www.amazing1.com/products/miniature-ferrite-high-frequency-transformer-2kv-10ma-20k-100khz.html

> 20k hertz
> 12v input

File: IMG_0554.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

how can i make this variable speed? I would imagine sticking a potentiometer beside the on/off but then again i dont know much about electronics.It only says 120V 60Hz 50Watts

>>950811
you would need a 50watt potentiometer
autotransformer/variac is better but bulky.
dimmer switches must be derated for inductive loads like motors before you ask

>>950813
what would a 50 watt potentiometer be in Ohms? they are rated on ohms. sorry as i said i dont know much.

>>950817

the amount of power that goes through a potentiometer has to do with the input power and the resistance because the resistance dictates how much current it will draw

>>950811

that motor looks about the same size as a fan, so i'd try a fan motor controller. of course, i have a couple of those already, so it wouldnt cost me anything to try it. you might be more hesitant to spend money on a solution that may not work.

also worth trying is just an old incandescent 600 Watt lamp dimmer. sometimes they work fine, other times you get weirdness, like hiccuping motors.

>>950784

with such a low current requirement, you can just hook it up to a 555 astable running at 12V. the astable formula in on Wikipedia's 555 page. the 555 is good up to about 600kHz.

>>950784

there's a PDF with schematics at the bottom of the page you linked to.

>>950904

That makes no sense to me so im just gonna go with an H-bridge over the transformer

Is it okay to use lineman's splice and heatshrink tubing for replacing a 800W heater cable? I had very little space so I didn't do the staggered trick, but it worked, I'm just wondering about the safety of it all.

>>950909
At 20 kHz you could probably use an audio bridge amplifier IC (TDA2030 or similar).

>>950954

yeah, it's ok, but next time drop a bit of solder on it to make sure it wont oxidize after a few decades. also, use a double layer of heatshrink.

File: 1439757356088.jpg (25KB, 403x403px) Image search: [Google]

25KB, 403x403px

How do I adjust the voltage and the amperage of an AC current as it suits me, provided that I have a DC power source should amplification be needed?
Is there a device specifically for that?

It has to be able to increase the amplitude of the AC when amperage is added (without producing any other distortions), and it should also be able to increase the voltage without affecting the amperage (and take the extra power from the DC power source if needed).

>>951150
dude do you even /ohm/?

>>951150
Not to be rude, but it seems by your question you don't really understand the fundamentals of electronics yet. I'd highly suggest reading some books in the OP before doing anything else. Don't want you to hurt yourself.

So Im thinking of using a surplus phone screen screen for a project, and Im wondering if I should just go with one of those ~7" screens they make for Raspberry Pi's and shit instead. Im just looking for a way to convert HDMI or DVI signal into something the screen can display.

Seems to be basically zero documentation for these things, despite how great they are (1440p 5" screen for $75 bucks? Fuck yes!) so Im even less certain that the one or two methods I've read about would even work with my model.

http://www.etradesupply.com/htc-one-m9-lcd-screen-and-digitizer-assembly-black-26135.html

Thats what Im hoping to use, though I know full 1440p is a lot to ask of whatever tiny board Ill have to use to convert the signal.

What can you guys tell me about circuit board pens? Don't work, too expensive, gave you cancer, best thing ever? More specifically, are they any good to use for fixing spots where things need to make physical contact with each other?

File: tmp_13966-20160223_121930-847846358.jpg (874KB, 2048x1536px) Image search: [Google]

874KB, 2048x1536px

Is this a legit way to measure the output current of a (homemade) stun gun?
I'm aware measuring the current in short circuit is not the same as measuring the current peaks that occur during arcing. I just want the value for comparative purposes.

>>951292
well not really.
air has a pretty good resistance so if you want to measure current of the spark you need to include the spark which will have a voltage drop.
if you want to measure current through a person then you need to include that resistance. pro-tip: human body resistance can be pretty much anything. if you chase a fat sweaty guy in the rain underwater and stun him the resistance will be very low. very very low. i know people who can touch mains live conductors (240v) to test them because they have fucked up skin.

>>947476
i attest It can work if the phone has usb otg. I used mine to upload sketches and do serial communication to a arduino with arduinodroid app.

>>947484
dude this is the internet everyone is an expert

>>948774
read this
>>950356

>>950430
constant spark is too much power for small batteries. Check air dielectric resistance for initial calculations

I recently received an arduino one as a gift. I'm a first year Electric Engineering student, any interesting project suggestion related to what I'm studying?

>>951512
get one of those 3pin thermometer ics and a logic mosfet, drive a 12v computer fan at variable speed off its input

>>950835
still have literally no idea.... i just want to know what potentiometer rating to get....

>as seen on 4chan

nes zapper and screen synth-anon got on hackaday

http://hackaday.com/2016/02/25/nes-light-gun-turned-video-synthesizer/

>>950430

You can replicate a taser circuit. I think you can use a relay as an oscillator to turn a coil on and off several times per second, producing the high voltage necessary for a spark.

>>952115
>tazer
>relay as oscillator
there is no upshot to using a relay as an oscillator what so ever.

>>952116
>not wanting your oscillator to make a deafening clacking noise

>>951907
When it comes to selecting passive components, there are several properties involved.

Foremost is the "value", i.e. the resistance of a resistor, capacitance of a capacitor, etc. In a theoretical model of the circuit, this is the only value which matters.

The other important properties are the limits: maximum current, maximum voltage, maximum power dissipation, maximum operating temperature.

(A third set of properties are the parasitic values: every component has resistance, inductance and capacitance; but you can largely ignore parasitics at low frequencies).

The kind of resistors which are used in low-current paths have a maximum power dissipation of a fraction of a watt. A 50-watt resistor is going to be physically large, and designed to be bolted to a heatsink. You will need to understand the thermal side of things in order to avoid setting anything on fire.

Having said all that: it's not practical to vary the speed of a 50-watt AC motor with a potentiometer. You either need an autotransformer or a variable-frequency drive (VFD), depending upon the type of motor (the speed of a synchronous motor is synchronised to the AC frequency, so if you want to change the speed you have to change both the frequency and voltage).

>>946499
>turn headlight switch on
>back lights for guages don't work
>replace fuse
>back lights and dimmer work perfectly
>, > 24 hours later notice back lights arnt working
>fuck with light dimming pot
>lights barely turn on for a second
>lights go back out
>fuse still good

>remove switch cluster
>5 contact pot
>pot looks like shit in some parts, looks new in others

Can a bad potentiometer cause a fuse to blow due to corrosion/dirt?

all that is in the switch cluster is 1 pot, 2 push bottons, 2 diodes, and a resistor.. a new one costs $100 and I'm not about to pay that much for the parts listed.

The circuit board is epoxied in the back of a shell, imagine a small black cup shaped like a retarded triangle football(hand egg), with the a circuit board that has only the back edge in contact with the bottom of the cup, with epox's deathgrip on the edge of this board.

What should I do about this situation? right now my thoughts are dump acetone in there and see if the epoxy will dissolve so that I can attempt repairs.
Or I could cut around the shell so that the epoxied part is cut out, but the problem with this is the connection in the back would get destroyed and require directly soldering the wires to the board, or gluing/taping the connection where it should go.

Im wanting to keep the original look so drilling a hole in the plastic front plate and putting in a new pot is out of the question.

local junkyard is run by a pissed off jew so that not an option either.

Say I wanted a 120 VAC to <24 VDC converter, and I need it to be quite powerful and also fairly light (say... >500W and <1 lb). Does anything like this exist? If not, could it be built? I understand it would take a transformer and a rectifier, but I have no real EE background and don't know what it'd take to determine it's power/current capacity.

who /gets components for their projects by asking for free samples/ here

>>952358
That's called a welder.

>>952395
do you have any advice about getting reference implementations for ICs?

>>952437
What do you mean? Like breakout/dev boards?

>>952441
I mean dev boards. I'm not enrolled as an EE student nor a practising EE just a hobbyist. maybe time and not education is the limiting factor, i.e. you can send a letter just after a new microprocessor is released

>>952448
I don't think any of the component manufacturers hands out dev boards, but you can ask for the chips on the boards just fine.

http://www.ladyada.net/library/procure/samples.html

>>952449
thanks for the resource. it''s a shame they don't do so anymore but published schematics help too

If I wanted to switch 25 volts with 5v output of a microcontroller with a darlington transistor, could I just put a voltage divider from the input of the transistor to the base so that it has enough voltage on the base to switch when I apply the driver signal?

>>952358
It's called a switched-mode power supply (SMPS).

The main difference compared to a traditional power supply is that it performs the conversion using high frequencies (from tens of kHz to a few MHz) which means that the transformer only needs to be a fraction of the size.

>> No.952505
Assuming that the load is between the 25V rail and the collector, the base only needs ~1.5 V.

If the low side of the load *needs* to be connected to ground, then you need to connect the base to 25V via a resistor and to ground via a NPN transistor which is switched by the uC.

>>952075
It wasn't the first time. This is the same anon.

http://hackaday.com/2015/12/05/a-sound-and-led-tastic-tricycle-shopping-cart/

>>946543
Fucking noice. Why can't these sorts of things be grad projects?

>>946983
late as fuck but VB is just a reference voltage of like 4.2ish volts because R9 and R8 make a voltage divider which is then coupled with C5 to make a steady voltage. VB is then used as a reference for all the op amps.

>>947661
The perfection of these lines is actually giving me an erection. Do you use a ruler or have you just trained a steady hand?

File: input.webm (1MB, 1280x720px) Image search: [Google]

1MB, 1280x720px

>>952616
It's roughly based on two projects I did back in school. My third year project was a light gun game, and my grad project was a sonar based touch screen.

I don't think I could have pulled this off as a grad project. I'm smarter now and I've got more time to work on projects. I'd also have to present it as a straightforward touch interface instead of a psychedelic art piece.

This is the light gun game:
https://www.youtube.com/watch?v=OMP28JZBSlo

>>952640
>cannida
xD

>>952505
I don't understand your question, but it appears that you have a 25V device you want to switch on and off using a transistor and a micro.

I/O pin of the micro goes to the base of the transistor. Emitter gets tied to ground. Put the load device between the 25V source and the collector.

>>952656
>>952505
(Assuming the transistor is NPN type)

Anyone here into etching their own PCBs? I want to do it, but I also want the boards to look pretty, with solder mask and silkscreening and whatnot, but I don't know how to do that shit cheaply.

>>952523
Thanks, looks promising.

>>952697
You can't really do it cheaply. The reason why companies can get away with it is because they do them in bulk. It'll never be more cost effective to do it yourself.

File: lm358.jpg (22KB, 600x344px) Image search: [Google]

22KB, 600x344px

What's your favourite op-amp IC?

Mine is the LM358: cheap, decent performance, works with single +5 V supply.

>>952721
You can make (black) silkscreen with the toner transfer method and solder mask using photoresist. Not exactly expensive or out of reach, but IMO not really worth the hassle either.

>>952766
My oil fryer uses an LM258, so probably that one c:

>>952237
>>952116
>wanting your oscilator to survice more than 10,000 cycles

>>952832
>not using eeprom as ram

>>952845
>not swapping OTPROMs at sanic speeds for maximum RAM performance

>>952832

> he fell for the "astable jew"
> he believes in magic crystals

i know nothing about driving mosfets in their linear region, but i'm interested in ldo regulators with fet pass elements. does anyone have good sources for actually designing fet ldo circuits? i mean design in the engineering sense, not bullshit physicist equations with variables that can't be derived from a datasheet.

I wanna get into electronics, where's the best place to start?

>>952977
buy an arduino, a breadboard, some jumper wires, and the shadiest looking cheap chinese multimeter you can find.

what sort of projects interest you? driving motors? pretty lights? wireless? lcd screens?

>>952979
Motors and screens mostly. At some point I'd like to set up a 3d printer and a CNC router, but that's pretty far off due to the whole fuckin' expensive thing. I think a raspberry pi-based emulation box would be a fun project too.

Why a cheap multimeter? Are they easy to fry if you don't know what you're doing? Solderless breadboard or proto board?

>>952981
i was joking about the multimeter, but its quality doesn't really matter for imprecise low voltage stuff like most arduino projects so don't waste cash if you're not sure you'll be really into diy electronics.

i'd also recommend a breadboard because you'll make mistakes, and desoldering is no fun.

as far as screens go, you can get nice <4" touchscreen breakout boards for less than $20 which will do a good job of introducing you to the boundless joys of microcontroller programming. if you're looking for a model, i used the hy28b in a project last year without issue.

i can't suggest any motor projects because motors need a thing to provide power to, so unless you've got something in mind they're rather boring.

Does anyone here have any experience with interfacing to an RTD? I need to measure temperature gradient in an oven and am trying to decide between using two sensors in a Wheatstone bridge (for each sector in the oven) versus constructing a four wire resistance measurement system with a constant current source and an instrumentation amplifier to measure the voltage drop across the RTD when applying a 350 uA current through the loop.

Anyone who has experience with RTD's what are some things that you experienced with your method of measurement? I would like to at least get accurate to ±5 °C.

>>952977
Reading the OP post.

>>952979
Arduino isn't electronics. It's just writing software that happens to interface with hardware

>but connecting the different boards is electronics!
Yeah, in the same way that putting pieces of lego together is "mechanical engineering"

>>953017
You're being a twat.
The arduino is a uC. There's nothing wrong with it for simple hobby electronics.

>>953021

Not the guy you're responding to but that doesn't even teach him basics about filters, op amps, shit like that. It's fun, but he won't know how to properly connect shit to it even.

>Mike is in the pastebin but Big Clive isn't
what

Big Clive and Mr Carlson's Lab should be added.

Also, what's the deal with Adafruit's LCD modules? Are they just shitty chinese modules marked up 250%?

>>952832
>>952845
>>952894
Fucking relays burn out you nigger.
I reply to TWO posts talking about relays.
How do you get eeprom, ram, and crystals out of this?

>>947173
Dave lives in Sydney. Closer than you think

>>953084

using EEPROM as RAM is like using a relay oscillator for a clock source

Crystals and astable timers are better alternative clock sources

>>948774
Please do not do number 1.

In these LED strips, the rails run all the way through. This way, you can cut them to any length and still have electrodes on each end. You can also feed from multiple locations to solve voltage drop.

With these long, common rails, each block of LEDs, usually three, completes this circuit with the proper resistors already installed. As more segments are added, they form many connections in parallel. Basically, a ladder.

With option 1, you are connecting the positive and negative rails directly to one another, shorting your supply. Connecting things labeled as positive and negative directly with a wire is never a good idea, but is occasionally forgiven by the gods of the magic white smoke.

Option 2 is the proper way to do this. My guess as to your observations is a bad supply. If you started with option 1, a soldered-in dead short sounds like a great way to make a good supply in to a bad one. If you haven't done 1 yet, you probably have a wimpy supply that can't push enough current to observably light the LEDs.

>tl;dr
>Option 1 is a dead short. Stop doing that immediately
>Option 2 is correct
>If problem persists, it's the supply. It either started wimpy or you fried the poor thing doing option 1

>>953035
op amps are ridiculously overrated, college students just worship them because that's what they learn in their useless EE classes and they have to be smug about something

you don't need op amps to make an automatic lightswitch or half the other simple shit new diyers do

>>949221

Not really, but a bit of redundancy always helps.

File: prahcessah.png (140KB, 407x350px) Image search: [Google]

140KB, 407x350px

Can anybody explain what does this circuit do?

>>953021
Arduino doesn't teach shit about power supply design, the physics behind electricity, etc. You literally don't even consider Ohm's law or LC reactance when you use an Arduino.

Sure, it may be uC, but it on its own does not encompass "electronics" as a whole field. Even in embedded systems design you have to figure out how DACs and ADCs work. How PWM works. Arduino doesn't accomplish this.

It's just a toy of a microcontroller dev platform.

Calling Arduino "electronics" is like calling lettuce a "sandwich". It is only part of a sandwich, and it on its own isn't a real learning experience.

>>953152
One of the Beatles said something like this when talking about stereo mixing: "It took way too long to figure out the drums go in the center".

For the first couple decades after stereo recordings became available the drums would be off to one side like how band members usually position themselves on stage. Eventually we figured out that music feels better when drums and other base sounds are at the same volume in both ears. Your circuit does that.

>>953152
spatial space processor circuit (makes the sound heard in the headphones sound like its coming from open air speakers)

>>952990
Your primary problem with Pt100 would be the DC offsets caused by your amplifiers and thermoelectric effects. The offset problem isn't going to be anywhere near 5C with good amplifiers and you can reduce it further by using Pt1000s. That said, AC measurement gets rid of the whole problem and is usually pretty easy to implement.

If you care only about the temperature difference and the difference is relatively small, the bridge with two sensors would probably give you better performance. You're likely to lose the advantage at higher temperature differences, though. And there will always be some faggot asking you what is the actual temperature and if you have to answer, you lose the remaining benefits of a bridge.

>>952990
Oh, and the need for 4-wire measurement depends on your wiring. Pt1000s would allow much higher wiring resistances.

File: eaglehelpconnections.png (57KB, 1082x730px) Image search: [Google]

57KB, 1082x730px

>>946499
Hey, /ohm/

So I'm "designing" my first PCB. I just want to control some LEDs and a couple of servomotors. I've downloaded Eagle and an instruction manual for it, but I'm stuck at a rather trivial thing. I can't connect the output pins of my 16Fx PIC series to the 4051N (de)multiplexer I want to use. When i use the net command I can't line up a wire going from one of the 4051N's pins to one of the PIC pins. When I move the PIC I see that no connections have been made. I really don't know the program well, so I'd appreciate it if someone could help me ...

>>953205
just deadbug it lol, what are you, some kinda pussy?

>>953205
Eagle is finicky that way. I don't remember exactly where you need to drop the wire end. Just play with it till you get it right.

I do some mild hobby electronics and I am trying to build a shitty surface using a raspberry pi 2, a laptop screen and a Anker E7 battery.

This battery says it can do 4A 5v output, and has three USB ports, I realised this is great for powering a raspberry pi but when it comes to powering my screen, which requires 4A @ 12V I might be in trouble.

Is it possible at all to get 4A @ 12V using two 5V 4A USB ports?

>>953211
Your screen needs 48W and your battery can supply 20W, so no. Well, there's a chance that your screen needs less than 4A and that your battery is willing to deliver more current, but it's still a really bad idea.

>>953220
Any reccomendations on powering the screen for "a really long time" without spending a bunch?

I was initially excited because I could theoretically get a lot of battery life but it seems that's not going to work out.

>>953222
Just get a different screen.

>>952656
>>952659

what I mean is that a darlington transistor will need more than 5v at the base

Basically, I want to take some voltage from the collector to help switch it on

So if it needs 12v to switch, I would put 7 on the base while it is off through a voltage divider from the collector

Where can I get free passives?

>>953389
college electronics labs

>>953389

thrift store

File: led.png (49KB, 1030x490px) Image search: [Google]

49KB, 1030x490px

Hi /ohm/

I'm looking for a cheap light sensor for a small weather station project. I experimented with an ordinary 5mm green LED by directing a beam of light into it and measuring the voltage of the LED. Can I conclude that the voltage signal is fairly linear from ~0..3000 lux?

>>953416
part number LTR-329ALS-01

http://www.us.liteon.com/opto.prdct.form.html

So I ``successfully" recabled my Beyer DT770s.
Turns out I fucked up and now the left driver a lot quieter than the right driver.
Left measures 92 ohms, right measures 91 ohms.
Electrically I'm pretty sure the connections are perfect.
Could I have possibly burnt the driver while soldering to the pins?

Interestingly, I have found while messing with the balance setting that "Left" actually equally bleeds into the right half.
"Right" just stays on the right half.

I tried to wire it in a 4-pin balanced configuration, and a bridge is extremely unlikely. The cable is electrically perfect, although I combine the grounds for the sake of using 3-pin 3.5mm.

Set your grid size to smaller. It's under view->grid.

> I'm "designing" my first PCB.
That's the schematic editor not the PCB editor. You can skip the schematic stage and go straight to the PCB if you like.

>>953373
Any reason you can't just use a regular-ass NPN transistor?

>So if it needs 12v to switch, I would put 7 on the base while it is off through a voltage divider from the collector
Voltages do not stack the way you think they do.

>>953373
> what I mean is that a darlington transistor will need more than 5v at the base
A single BJT needs 0.65V. A darlington pair needs 1.3V.

The only way that you need more than 5V is if the transistor is on the high side of the load (i.e. the load is between the emitter and ground). You'd normally put the load between the collector and the positive rail with the emitter connected to ground.

If you need to have the load on the low side (e.g. because one connection needs to be grounded for some reason), the easiest approach is pic related.

File: level.png (3KB, 461x518px) Image search: [Google]

3KB, 461x518px

>>953530
Damn, forgot the pic.

File: level2.png (4KB, 513x496px) Image search: [Google]

4KB, 513x496px

>>953530
And if you absolutely must use that NPN darlington, then you can drive it like this. But note that the output will be inverted, i.e. a logic 0 from the uC will turn it on, logic 1 will turn it off.

File: level3.png (7KB, 756x596px) Image search: [Google]

7KB, 756x596px

>>953533
Finally, if you need to drive a grounded load through an NPN switch with non-inverted logic, there's this.

If nothing else, this should help make it clear why the load goes on the high side wherever possible.

File: 09387454.jpg (352KB, 1000x562px) Image search: [Google]

352KB, 1000x562px

Where does /ohm/ get their wire from?

I used to buy this wire from a local electronics store, It's perfect for my needs because 4 wires fit into a piece of paracord, it's thick wire with thinner insulation, and it's not a solid core. But they stopped selling it and they don't have anything that fits those criteria anymore

OD of the insulation - 1.2mm
OD of all the metal strands together - 0.8mm

It used to come as 6 wires inside a larger insulator and I'm pretty sure it was for security cameras.

File: headphone.png (8KB, 947x506px) Image search: [Google]

8KB, 947x506px

>>953152

>>953530

My darlington transistor is high power and it needs more than 5v

Also, in those schematics did you actually use 2 transistors to represent 1 darlington array or 1 transistor to represent 1 darlington array

>>953603
Darlington arrays are always represented as 2 transistors.

>>953608

Mine is just one package with 3 pins. not sure about internal connections.

>>953636
Are you sure you even know that what you have is what you think it is?

Just noticed that if i hold my AC volt stick (pic related) in my left hand and my phone that's plugged into its charger in my right hand (or if i just touch them together), the volt stick gets a reading. Does this mean that there's current flowing through me?
It's an ac volt stick, doesn't the current get rectified to dc?
I'm not great with electronics, just thought this seemed a bit strange

File: images (6).jpg (8KB, 384x384px) Image search: [Google]

8KB, 384x384px

>>953669
Forgot pic