/ohm/ - electronics general

last thread:
>>1125052

How do you guys usually deal with enclosures

I'm trying to build a simple power supply and am having problems finding things to put it in

I'm scared I didn't solder my heatsink to my cpu like op's pic
how screwed am I?

>>1133322
you need to do it right now. you're extremely lucky you're able to type this.

electronics have a phenomenon called "leakage current" where the electrons leak out with time and eventually they stop working. if your heatsink isn't joined fully to your cpu then you've got electrons leaking into the air right now.

>>1133310
gut a junk atx supply. presto you have a metal box with a integrated fan

File: original.png (4MB, 3264x2448px) Image search: [Google]

4MB, 3264x2448px

Circuit bending an old Yamaha PSS 30
and the speaker has a very high pitched noise constantly coming from it.
any advice?
if I find a bend that silences it should I just wire it so that it stops?

File: 023983083.jpg (852KB, 1080x1080px) Image search: [Google]

852KB, 1080x1080px

>>1132721
Turned out alright.

>>1133263
>pic
what is he thinking about?

>>1133399
Watermelon.

>>1133263
Great pics!

>>1133399

>why is the photographer asking me to hold the hot part?
>why is the base station pointed away from me?
>why is it off?
>why is there an open harddrive?

>>1133372
nice job

File: 20151011_163235.jpg (1MB, 2048x1536px) Image search: [Google]

1MB, 2048x1536px

>>1133367

Sounds like you need a filter somewhere.

>>1133399

>phew! So glad I can take a break doing my normal duties as chief rocket scientist at NASA so I can do a modeling shoot.

>>1133423
>Why are there no height markings and why am i holding a soldering iron instead of an inmate number?

>>1133263
If I salvaged power supplies from old appliances, could I build a high voltage DC power supply off of them? Looking to hit 9-10 kV range

>>1133399
probably realized the limitations of silicon based chips and went into deep thought on quantum computing

is his head dis-proportional to his body?

>>1133460
if you're talking about computer psus you'll want a flyback transformer specifically. you might get away with using the secondary winding as your primary for stepup but if its turncount is too low for your input waveform then you'll have to rewind it. after the flyback you'll need a multi stage voltage multiplier whose diodes you might not be able to salvage. you'd honestly be better off getting your shit off digi. it won't be too expensive as long as your multiplier stage isn't putting more than 1kv stress on any diode or cap.

you can also just stack (or rewind, gl lol) microwave transformers but be aware that they operate slightly into saturation and will heat up if run continuously.

File: 20170219_114737_kindlephoto-365793.jpg (377KB, 1432x1452px) Image search: [Google]

377KB, 1432x1452px

I was the guy from:
>>1133230
>>1133199
>>1133196
>>1133189
>>1133099
>>1133044
>>1132973

Shielding the cables was indeed the hot ticket. Suuuuper excited my $25k+ education barely mentioned this at all.

Note to self, making my own cables is a PITA (pic related). Is there any foil or braid wire out there with just one conductor? I never used the other conductor in my diy cables and I'd like to cut down on bulk/clutter with the final build.

Running my circuit bent stuff through it, seems the low power instruments (2xAA) get drained in the 'off' positions as the LEDs on them go dark even when pressing buttons. Not a big deal.

However, some have contact points will still bleed through in 'off' mode. I've found tying these points to the input ground on my mixer rectified this, but is there a more practical way to go about retro-balancing unbalanced electronics (assuming this is a balancing issue)? I thought buffers might be the answer but I don't see how that would ground the contact points.

Also, is it better to connect all the shield lines to chassis? In pic related, the 'chassis' was cardboard so all shield lines were strung together ( and eventually to ground on in+out).

File: man-testing-equipment-with-oscilloscope.jpg (93KB, 1000x709px) Image search: [Google]

93KB, 1000x709px

What equipment do I need more for everyday electrical engineering
I have:
General purpose meter (AC/DC voltage, current and resistance)

Capacitance meter

A few test lights

Shitload of resistors, capacitors, potentiometers, diodes, wires ...

Soldering iron along with equipment

Raspberry Pi 2

I plan on buying an old Soviet oscilloscope for ~20$

What more do I need ? I haven't really felt I am missing anything but a power supply How much could a good power supply cost (perhaps another Soviet old ones, their equipments seems to be very sturdy and cheap right now)

Are the diodes in rectifier diode bridge packages reasonably well matched? At least any better than four random separate diodes?

>>1133571
While it is quite likely, your typical bridge maker does not give you any guarantees of the matching.
If you want matched diodes, get matched diodes. CA3019 was a well-known matched diode array, but there are others (less antiquated) as well.

>>1133399
>This is stupid but it's gonna pay my rent and tuition for the month and buy my imouto a birthday present

File: elenco-resistor-substitution-box-rs-500-13.jpg (90KB, 452x500px) Image search: [Google]

90KB, 452x500px

>>1133525
Could come in handy

>>1133525

If you're just going to do digital shit don't buy analog tools

anyone do any battery powered iot projects? i can configure a 433mhz receiver to sample at low rates but i'm not sure what's available on the market in terms of iot products to interface it with.

>>1133633
I don't just do digital shit, I use Raspberry for all sort of different things + digital and analog don't exclude each other

>>1133624
That looks nice and affordable too!
Will definitely be adding to my arsenal.

>>1133506
>Is there any foil or braid wire out there with just one conductor?
I use RG165 - small, flexible, stranded center conductor.

>>1133506
>Also, is it better to connect all the shield lines to chassis?

Only at one point - lookup 'ground loop' relative to electronics.

While you're looking up stuff, lookup 'lead dress'

>>1133635
ESP8266 is a really cheap WiFi module that can be used on battery if you're careful with using the power management to stay off most of the time. https://openhomeautomation.net/esp8266-battery/

>>1133750
thanks for the link, that's actually really interesting. i was coming up with 30uW for the 433mhz receiver sampling but i can live with 200uW for the convenience of just polling a server.

>>1133263
How screwed am I if I'm not holding the soldering iron by the hot metal bit like in OP's pic?

>>1133773
It would be extremely painful.

File: IMG_0817.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1133506
Coax/RF cable comes with built-in shielding I think, but it might be pretty inflexible. Your other bet is specialty audio cables, though you pay a premium for those. You're looking for "screened". I just buy shitty hookup wire and at least it doesn't look like an eldritch abortion.

If you use an Aluminium project box then you have an easy option for tying things to ground, otherwise you could try crimping big eye connectors to the shielding wires and screwing them together with the panel-mount sockets. With my first multi-socket audio hub (pic related), the face was made of a thin sheet of aluminium (torn off a heatsink), so I didn't have to spare any wire to hook up the grounds, so a conductive chassis will certainly reduce clutter. It will also stop you picking up RF, if you can even have that problem.

Your draining battery problem sounds like a faulty connection in the instrument's on/off switch, if it has one. If not, install.

If you think balancing is the issue, it sounds like you might need a balun, whatever that is.

>>1133364
How about for items bigger than those?

I've been trying to communicate with a small i2c oled module (ssd1306 is the controller) using my tiva mcu, but I keep getting arbitration errors. It works just fine on an arduino, but when I try to send the exact same bytes with the tiva, even with external pull up resistors, I can't get anything.
Anyone have ideas/experience with this?
Going to bring it to an oscilloscope on campus labs today or tomorrow to try to figure out what's going wrong.

>>1133933
The usual problems would be start/repeated start/stop/ack handling and too high speed.

File: ssd1306.jpg (1MB, 2051x2141px) Image search: [Google]

1MB, 2051x2141px

Getting my head around LCD panels for fun and because I wanna do some wearable shit.
Anyway, I'm seeing LCD/oled modules on breakout boards, I always assumed the driver was on the breakout board and every pin on the ribbon cable was basically used, now I'm looking at them and for most modules they just tie 90% of the pins to ground through a cap/resistor, and that maybe there's two or three pins for data, be it i2c or spi.
This confused the living hell out of me for a bit, then I realized the controller chip is the long (Maybe 1x20mm for example) die mounted on the bottom of the flex cable. Is this accurate?

File: New-LCD-Display-Screen-Replacement-For-iPod-Nano.jpg (13KB, 400x400px) Image search: [Google]

13KB, 400x400px

LCD'S ALL ROUND!
Let's say I wanna build a bunch things based off 6th/7th gen iPod nano LCD's. Short of just googling "iPod nano LCD connector" and hoping for the best, what's a good way to work out exactly what sort of connector this display here uses, so I can get a bunch of the female pair for my custom PCB's, and how could I go about working out the pinout? If it's custom apple I could half just probe at a running iPod till I work out what each pin does, but rather than that is there any decent way?

File: diy.png (1MB, 720x758px) Image search: [Google]

1MB, 720x758px

>>1133525
resistance decade
this one is homemade, cost me about as much as resistors themself. you can buy already finished ones like this i think. thou you will love it if you do it yourself

>>1133823
In my old job I used to just make one off laser cut boxes. Now I have no laser cutter so if it can't fit in my 3d printer's build area it's gonna be fucking raw.

>>1133773
Do you smell chicken or pork? If you do you're fine.

>>1133969
>wearable shit.
>shit
operative word.

>>1133989
Never seen one of those, it works by adding serial resistors right ? How exactly you add resistors on your board, it's kind of blurry, you use jumper cables ?

>>1133775
You're a big guy

>>1133506
Looks like helen keller took up bomb making

Not sure if here or stupid questions is better, let me know.


I want to build my own cables with woven sleeves. Only a few lengths of 2.1mm extension/pigtail, so nothing complicated.

I was just looking for anyone who had done something like this before and had tips on making it look neat+practical.

Materials:
Low gauge (18) stranded silicone wire
Paracord or "sleeving" (http://www.ebay.com/itm/6MM-1-4-TIGHT-Braided-PET-Expandable-Sleeving-Cable-Wire-Sheath-2M-5M-10M-/220785437600?var=&hash=item3367d66ba0)
Solderable 2.1 terminals
Some shrinkwrap
???
I havent purchased anything yet pls advice.

>>1133525
Where do you get old soviet stuff?

File: Untitled.png (554KB, 526x599px) Image search: [Google]

554KB, 526x599px

>>1134144
You have to live in a ex-communist country(or get very lucky on ebay), old soviet electronics are still around and you can easily find spare parts. I might have overestimated , i would need at least 50-60$ for a solid functioning oscilo, but I had a friend who bought a preserved one(that doesn't function) for 20$ and just changed one capacitor and the thing worked

like this:
http://www.kupujemprodajem.com/Iskra-osciloskop-42371160-oglas.htm
http://www.kupujemprodajem.com/Osciloskop-42967726-oglas.htm
http://www.kupujemprodajem.com/Osciloskop-Orion-1534-42027828-oglas.htm
http://www.kupujemprodajem.com/Prodajem-osciloskop-Ruski-S1-101-42942158-oglas.htm

File: server_4U.jpg (173KB, 640x450px) Image search: [Google]

173KB, 640x450px

>>1134074
4u

>>1134151
for 60$ you can get a modern 2ch analog one

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

2MB, 3264x2448px

I made this amp tonight and I still have to set up the enclosure for it. Is there a good way to secure the board to the enclosure?

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

2MB, 3264x2448px

>>1134204
Better pic with enclosure behind it.

>>1134204
>secure the board to the enclosure?

double sticky foam tape
or
standoffs to corner holes

>>1134151
In New Zealand, some guy made a sire called TradeMe before Ebay got big, so we have a seperate market for old vintage electronics there. Every time I look, I see old oscilloscopes, sometimes broken, sometimes covered in dust, they usually don't go for more than 80 NZD (that's like 60 USD), and one (slightly badly focused) even went for 8 USD, 2 channels and everything. Feels good to be a Kiwi.

File: Screen Shot 2017-02-21 at 9.41.04 PM.png (1MB, 1266x1168px) Image search: [Google]

1MB, 1266x1168px

>>1134244
This is what we've got today.

File: BLDC motor driver.png (12KB, 709x408px) Image search: [Google]

12KB, 709x408px

So I posted in the last thread about this problem
>>1132491
but I still can't really seem to figure it out.

>Trying to spin a 3 Phase BLDC from a HDD
>Pic related is circuit I devised
>Using a microcontroller (8051) to generate PWM
>Wired up on breadboard; motor doesn't move and my external power (10V source) supply shows high current and low voltage (5.07A and 1.7V)

I've tried replacing the P-FETs with N-FETs; no change
I've tried lowering the PWM signal frequency to something as low as 2Hz; no change
I've tried just hooking up the gates straight to 5V DC; no change
I've tried killing the power to the gates entirely; no change

To me, something is shorting out, but for the life of me I can't figure it out.

>>1134262
Update:

I tried disconnecting my gates from the pwm signal and hooked up my pwm signals to 3 respective LEDs. They behave as expected; light up sequentially based on the pwm freqency.

Now, when I try attach my gate terminals back to their respective pwm terminals... the LEDs no longer blink in succession and just all stay on permanently. I tested this with an ohm meter too; when disconnected from my mosfet circuit, the pwm provides a voltage of about 3.4VDC coming out of the pwm terminals and when i connect my circuit back to the pwm, my ohm meter reads 1.2VDC. So obviously my mosfet circuit is doing something weird and probably shorting something out. All my mosfets are freshly bought mosfets and I double checked my connections... everything is connected exactly like I did in my circuit.... is there something wrong with the circuit design?

>>1134245
Exactly, I live in Serbia (ex-Yugoslavia), and apart from Soviet electronics we had really good electronics factories here too, and many people still have random equipment in their basement that they inherited from parents/grandparents. At the time, things were build to last, so you can fix them simply, and if you spend a month checking oscilos on buy/sell websites, you will 100% find one for less than 10$. People don't know how much these things are worth and they are essentially giving them away. Now I am just waiting for a good 2-channel one that works and is under 60$

File: decade final.jpg (569KB, 719x2440px) Image search: [Google]

569KB, 719x2440px

>>1134061
Camera is just shit, nothing i can do about that sadly.
I did it by bending resistor legs down and onto the next resistor. That isn't really great, but it's good enough for 5V i'm using it on.

I am not aware oh how well do you understand electronics, so i'll explain as if you are a full on beginner.

The principle is shown on picture 3.
You have 10 positions for a jumper (i have drawn 5 of them (the blue color)), which allow you to set from 0 to 9 times the resistance value of a single resistor used.

Then on picture 4 you see how to add "banks" of resistors in series. You can set from 1 to 9 ohm on the bottom one, 0 to 90 ohm on the middle one and so on. My version goes from 1 ohm (or zero if o short all the bus bars) to 999 999 ohm.


If you are making this, here are some tips:

1.You can pick different power ratings for different resistors; i got (i think) 500mW for 1 and 10 ohm, 250mW for 100 ohm and 125 mW for the rest. Saves you even more money.

2.Don't bother getting more accurate resistors, because you can just plug in your ohmmeter (which is probably inaccurate as fuck anyways) and adjust the resistance value. I found that setting 540 ohms gave me 537 ohms, so i just added 3 more ohms from the 1 ohm decade

Apologies for the crude drawing, i set aside 5 minutes max for this (ended up taking 15 ofcourse).

File: H_Bridge3.jpg (29KB, 800x430px) Image search: [Google]

29KB, 800x430px

>>1134262

judging by this H bridge, you need to swap the position of the mosfets
(reversed polarity)
also, many use 1k to 10 resistors to gates

>>1134284
>the pwm provides a voltage of about 3.4VDC coming out of the pwm terminals
Not familliar with MOSFETs really, but I think that both 0V and 5V would turn on the top MOSFET, becaure to it, it's -5 and -10 volts, so they're always on. Check the voltage across one of them. is it near 0 whenever The PWM wire is connected?

Maybe a level switcher would help.
Feed the pwm wire through a 4.7K resistor to the base of a 3904 switching transistor.
Tie the collector of the 3904 to +10V through a 4.7K resistor.
Tie the emitter to ground.
Take the signal from the 3904 collector to the base of the MOSFET.

This will convert the 0-5V signal to 10-0V (inverted, so program accordingly)

>>1134336
>>1134337
I think the P-Fets were just constantly on which is why it looked like something was shorting out.

Anyway, I rewired the P-Fets to reflect the H-bridge in >>1134336 . I tried it out and my motor was jittering in on place while drawing 2A 6V from the power supply. I then tried rearranging the motor terminal wires in all different arrangements (ABC, ACB, BAC, etc) and it behaved the same in all cases. I then realized one of my LEDs that was monitoring the PWM signals was off. After some quick trouble shooting, I figured out one of my N-Fets crapped out and was just grounding the pwm signal. Swapped out that N-Fet, and now the motor won't move and will only draw 0.01A and 10V on intervals. In the midst of troubleshooting I shorted out my microcontroller so the thing is fried now (I am in the process of replacing it now).

Anyway, any idea why the motor was jittering and drawing so much current? Did I fry some other mosfets in the process?

File: key_fob_detector.jpg (16KB, 349x263px) Image search: [Google]

16KB, 349x263px

Are 20k and 22k resistors generally interchangeable? I'm building a basic key fob tester and the schematics calls for a 22k resistor but all I got is 20k.

>>1134366
Never mind, find a packet of 22k.

>>1133310
OKW enclosures has a good sample system, you can get all their stuff for free

File: Screenshot_1.png (2MB, 3839x1079px) Image search: [Google]

2MB, 3839x1079px

hey diy So i wanna make a “massage” custom toy with bluetooth and all the parts have a very small place to fit so i already got an hc-06 bluetooth module and i was thinking for taking these to complete it.
Q1 should i get a 3.3 or a 5v arduino mini.
and most importantly Q2 how do i power it and what kind of battery should i use. I want it to be able to run at max for 10-15 minutes and be rechargeable.

I am only getting the adapter to pass the code after it has it then its not gonna be part of the final build.

File: asdf.png (54KB, 1570x653px) Image search: [Google]

54KB, 1570x653px

I'm a beginner in electronics and I have a problem with the transistor from a kit I assembled.
Sometimes T1 (highlighted in the schematic) self activates for some reason, when the output 10 of U1C is 0V. It happens randomly whenever I touch S3 or if I touch the base of T1 with a screwdriver (unconnected with anything else). When it happens, D3 is halfway lit (it's brightness varies) and the voltage between the base and emitter of T1 is 0.1-0.5V.
What could be causing this?

>>1134496
Sounds like a bad solder joint somewhere between U1C and T1.
It is also pretty normal that touching the base turns T1 somewhat on, particularly when R5 is large. You can try smaller R5.

>>1134498
>particularly when R5 is large
Why is that? What's responsible for it?

>>1133263
The guy's head is too small for his body.

>>1134424
Q1: use a 3.3v, it will be easier to interface it with the HC-06
Q2: use 3x18650 li-ion batteries (buy 3 battery holders from Aliexpress), you can recharge them individually with that TP4056 you posted. Also, use a dc-dc step down converter to bring 12.6v to 3.3v and switch the motor with either a bjt transistor or a logic-level mosfet

>>1133496
Yeah, I think I will just swipe a flyback generator from a microwave. Was gonna pull one off of an old CRT TV but after reading how they function, fuck that.

>>1133263
hello, retard here; if i have a bulb marked 12v 3.4 amp, the pertinent info is the voltage, right? so i could hook it up to a car battery by itself without any problems?

>>1134570
Yes
Hook it to 12v, and it will draw 3.4 amps

>>1134502
There is electricity everywhere, travelling through the air. A lot of it comes from the power lines but there is also radio, wifi, microwave background, lightning, etc. When you have a high impedance input any large piece of metal acts as an antenna and may overpower the existing signal. With a smaller resistor the "good" signal will be stronger so it will drown out any interference. Of course that can have other consequences so you always have to find a balance.

File: ground lift hypotheticals.jpg (156KB, 800x1329px) Image search: [Google]

156KB, 800x1329px

>>1133506
>>1133663
>>1133782

Looking up more info on ground loops, specifically how to implement ground lift switches: the only diagrams I come across concern DI boxes and I'm not working with XLR at all.

So pic related: I'm trying to figure out how to put a ground lift switch at the input(s) of a build. I included two different ways to go about it, let's call top part diagram A and bottom part diagram B.

My question is, will either part of pic related work, and if not, what should I do so that it does work?
Also, if I have multiple inputs, should they have their own ground lift switches or can they all be run to one ground lift switch?

How much power is safe to dissipate on 2w resistor with passive cooling?

>>1134631

2w?

>>1134631
The usual rule of thumb is half of the rating.

The exact answer can be found from the resistor's datasheet and it varies somewhat between different makers and types. For through hole resistors it is common that the rated 2W applies at 25C temperature, in free air and with untrimmed leads. The resistor will also run really fucking hot.
SMD resistor power ratings are often for 70C ambient temperature, but in their case the PCB routing is a big part of the heat sink. You won't be able to dissipate the specified power unless you pay attention to the wiring.

>>1134608
>When you have a high impedance input any large piece of metal acts as an antenna and may overpower the existing signal
This is the part I don't get. Why is that? Is noise behaving like a current source?
Could you point me to some resource from which I can learn more about it?

>>1134617
Use isolating transformer you knobber.
Fucking plastic washers what are you doing m8.

>>1134747
not really an antenna, but high Z + temperature

https://en.wikipedia.org/wiki/Johnson%E2%80%93Nyquist_noise

>>1134747
>Is noise behaving like a current source
You can see it as a current source. Or as a voltage source or as a power source. Ultimately it is about the power which can drive your transistor. The base resistor bleeds part of that noise power away, reducing the amount your transistor sees.
If you have access to an oscilloscope, just try sticking the probe to various pieces of metal and to yourself. You can also try connecting the probe tip to the ground clip via a resistor and comparing it to the results without it.

I don't really have any good resources, but at least you can try these:
https://en.wikipedia.org/wiki/Electromagnetic_compatibility
https://en.wikipedia.org/wiki/Electromagnetic_interference

>>1134751
That's different kind of noise and many orders of magnitude less than what you can couple to your circuit by touching it.

File: Latest Gaming Gadgets_ USB Beverage Cooler like refrigerator(1).jpg (22KB, 400x556px) Image search: [Google]

22KB, 400x556px

I'm playing around with the peltier from one of these, but its only acting like a 5w.
I was wondering what sort of strife I'd get in if I hooked it up to 12v instead of 5?
I'm really trying to get this thing to freeze, keeping the hot side cooled constantly.

>>1134817
peltiers are resistive, their voltage limit is governed by their maximum power dissipation (they're 50% efficient at best). yours can probably handle more than what it's pulling from usb though.

i've already looked at fast cooling drinks and you'll need better thermal coupling than one of those shitty boxes offers anyway.

File: 1.jpg (67KB, 1000x1000px) Image search: [Google]

67KB, 1000x1000px

>>1133263
If I input 19v into a car charger (steps down 12-24v to 5v) would it be fine?
Would it still output 5v?

>>1134747
https://www.extremetech.com/extreme/148247-german-student-creates-electromagnetic-harvester-that-gathers-free-electricity-from-thin-air

>>1134918
You just said it steps 12-24v to 5v.

>>1134617
ground loops are beyond me, but I'm pretty sure that both of those sockets will be attached to the chassis in pic 2 regardless of the position of the switch or the insulation properties of the washer, because the two grounds are wired together.

>>1134971
I'm just not sure because cars are 12v and trucks are 24v, but what if the number isn't 12 or 24?
Cars and trucks don't operate outside those voltages

Would it step up 12-24 and everything in between, or only 12v and 24v?

>>1133506
i need to puke

>>1134918
>(steps down 12-24v to 5v)
>>1134991
>but what if the number isn't 12 or 24?

It's for 12-24 (twelve to twenty four)

But from what you've said so far I'm betting you'll manage to hook it up backwards and fry it.

File: Voltage-divider-circuit-with-half-the-voltage-breaboard-schematic.png (39KB, 488x308px) Image search: [Google]

39KB, 488x308px

Stupid question:

So voltage dividers...

Why is the voltage output in the middle of the 2 resistors? Would it make a difference if it was before or after?

>>1135041
Voltage from top of top resistor to GND:
>Vin volts (red wire)
Voltage from between the resistors to GND:
>Vin*(Rbottom/(Rtop+Rbottom))
Voltage from bottom of bottom resistor to GND:
>0 volts (black wire)

>>1135041
It's all relative, so you're taking the voltage between the centre and one side.

The voltage across both resistors is the maximum, say 5V. This means that 2.5V is across the first resistor, and 2.5V is across the second. If you put a low load/high resistance (compared to the load of the resistors) between one side and the centre, the voltage across it will be about 2.5V. These circuits are necessary when using OpAmps, because they take power across the whole battery but require a virtual ground for balance purposes, and since the power is being drawn equally from one side as it is from the other, the circuit works fine. However, If you attempt to draw high current from just one of the sides of the voltage divider, the centre voltage will shift.

Say both of the resistors on the 5V circuit have a resistance of 1kΩ. If your appliance attached to a 2.5V division is 100kΩ, then the effective resistance of that side will be ~990Ω, so the voltage will be divided into 2.49V across the appliance's side and 2.51V across the other, meaning we made a minimal change. But if a 10Ω appliance was put across a 2.5V division, then the effective resistance of that side will be only ~9.9Ω, so the voltage will be divided into 0.049V across the appliance's side and 4.951V across the other. In this extreme case we're essentially short-circuiting the appliance-side 1k resistor and putting almost the full 5V across the other.

As you might imagine, voltage dividers of this design do bleed power constantly when on, so they aren't terribly efficient, especially for high drain applications.

I got no clue what happens if you exceed the battery/power pack's ability to provide current; it'll probably just explode in your face.

File: electronics-engineering-challenges.png (150KB, 1920x1080px) Image search: [Google]

150KB, 1920x1080px

hey /ohm/ im an EE student just now taking my first circuits class. thinking of doing some of these in the summer - which of them can be done relatively cheaply, at my level, and would look okay on a resume

>>1135075
97, 90, or 71, because they all have potential hippy "I care about the environment" connotations.

Also rolling.

>>1135080
HA, I already planned to build an FM jammer, close enough.

File: frog-hmm.jpg (105KB, 900x900px) Image search: [Google]

105KB, 900x900px

>>1135080
those are possible from just my first class?! how many others are? almost the whole thing?

>>1135075
Rolling for myself

>>1135082
No, I was just responding to the "good on a resume" part. The rest will all just make it look like you're autistic on a resume. For something easy, go for an FM radio, PSU, clock, useless machine, or maybe a Jacob's Ladder if you don't mind HV.

In my experience, once you've got V=IR and P=IV down, along with maybe the curves of diodes, inductors, and capacitors, mutual inductance, transistors and logic gates, you will be able to learn as you go. I built an Elektrosluch, and I'd encourage you to do the same if you want to learn about audio and opamps, and if you're cheap it will cost less than $50. (But I bought 100uF bipolar electrolytic caps for $6 each just to increase the bass, no ragrets.

>>1135089
woah, pretty cool. okay, so looks like once i hit summer and read up a bit i'll work on it. projects would be fun. by resume, i meant something more like a portfolio of projects for an internship - i'm hoping an FM radio doesnt look too autistic in that regard

These days, are there any advantages of compressor cooling over peltiers?
I've tried looking it up but all I see is shill articles by Peltier manufacturers.

Perhaps that you can just run a fridges pipes in open air to cool them down, when most Peltier systems need active cooling?

File: 1460204940033.jpg (4MB, 5312x2988px) Image search: [Google]

4MB, 5312x2988px

ch1 is rx
ch2 is tx

anyone have any advice to eliminating the noise shown on my 315mhz receiver? sorry for the phone pic, not digging up another usb cable.

File: 1475085971652.jpg (168KB, 800x566px) Image search: [Google]

168KB, 800x566px

>>1135103
pic related are shitty ebay transmitter/reciever

>>1135103
You don't eliminate it, you live with it. That is, you design your protocol so that it has an appropriate preamble and checksum and then you check each received potential message for correctness.
In principle you could tinker with the receiver to make it less sensitive, or better, find the raw signal and check its strength to discriminate noise.

>>1133975
mike s electric stuff reverse engineered one of thos check it out on his sit he even has sowtware i think

File: IMG_0820.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1135095
I think the problem with Peltier coolers is that the cold side is very close to the hot side, and there are losses there, and also the wire conducts heat between the two sides.

>>1135092
Here's it in a box. I don't think radios are too autistic compared to an etch-a-sketch robot, but that might just be me having a radio fetish. (I'm currently winding an inductor to wire to a 3.5mm mono jack to transmit audio waves as radio waves so I can listen to them wirelessly from the elektrosluch while simultaneously producing strong radio pollution)

>>1135075
Roll

I have one fan with a mx input current of 0,12A, max input power of 1,44W and 12V voltage, if i buy a usb to 4 pin connector, will it be sending too much current or power to fan? Also, will the voltage difference make it spin slower or something?

>>1135317
its not even going to turn at 5V

it could run but you'd have to give it a kickstart every time

File: -1366594430-1905410607.jpg (98KB, 678x675px) Image search: [Google]

98KB, 678x675px

>>1135317
Get one of those little 5v to 12v step up adapters from aliexpress. 1.5w is easy enough for a USB port to provide, even from a computer.

MT3608 DC-DC Adjustable Boost Module 2A Boost Plate 2A Step Up Module with MICRO USB 2V - 24V to 5V 9V 12V 28V LM2577
http://s.aliexpress.com/qqA7fEv2

>>1135320
>>1135325

Thanks to both, i might get the adapter then.

>>1135325
>>1135320
Also, i want to be able to control the rpm, to i just have to add a potentiometer? Of how many Ohms?

>>1133263
Does anyone know if Instructable's new electronics/robotics classes are any good for a beginner?

>>1135402
>i just have to add a potentiometer?
The blue square on the module is a potentiometer.

>>1135418
Link to the page: https://www.instructables.com/class/Electronics-Class/

>>1133525
hot air station
microscope
smd soldering station

lol no you're fine. a hot air station is nice but not necessary.

>>1135419
That one isn't good for what i want, i wil need to adjust it any time i want, is there a knob for that type?

>woman
>working on electronics

lol, it's like when you give an infant one of those fake steering wheels

>>1135450
Unsolder blue pot from board, measure, replace with the type you want of the same value.

>>1135474
Ok then, thanks

>>1135457
There are a few autistic girls majoring in EE at my Uni. Most of the cute ones switched majors.

>>1134918
It'll work, but might get very hot depending on your load. Those cheapos usually have a linear regulator inside.

I have a pretty basic Arduino car model with 2 50mm radius wheels on one 90mm length axle, where each wheel is powered by a continuous servo

For forward motion, I would model each servo as being set to 180 for its max speed. How could I use an acceleratometer to then determine the velocity of the car? Does it take n ms for a continuous servo to spin to its max RPM? How could I find that out?

File: BLDC motor driver - Output Issue.png (40KB, 748x1228px) Image search: [Google]

40KB, 748x1228px

I have tested my input and output response from my circuit using LEDs. I have included some crappy graphs in pic related to portray what I've observed. I want the output to have the same behavior as the input (controlling for a 3 phase BLDC posted earlier >>1134262).

I was using LEDs to see how my input and output was behaving. Interestingly, I get different behaviors depending on how many of the PWM inputs I have an LED connected in parallel.

>0 LEDs connected; no output on ABC
>1 LED connected (one of a, b,or c); very weak uniform output on ABC
>2 LEDs connected; strong uniform output on ABC
>3 LEDs connected; strong uniform output but no pulsing on ABC

Something is going on, but I have no idea what. Obviously putting LEDs in parallel with the PWM signal modifies the ouput behavior on terminals ABC.

>>1135509
That shit's whack, you should get as many different diodes as possible and test different arrays of them, measure their collective voltage drops, and find the relationship between diode drop and output waveform, and then see if you can find a theory that matches with the behaviour of the FETs.

Is it a bad idea to buy a Maxim IC off aliexpress?

It's much cheaper than other online vendors.

>>1135567

yes, you never know what you're getting when you buy chink shit. see this thread:
>>1135558

>>1135509
Is your driver ground connected to your uC ground?

>>1135474
I will still need to adjust it from 0 to 12V, i don't want to blindly adjust it and burn the fan because i was on 20V without knowing.

File: DSC_0275.jpg (1MB, 3133x3533px) Image search: [Google]

1MB, 3133x3533px

Stupid question, what are these? I found them in a box that I ordered orange LEDs in, are they resistors?

>>1135527
Could give it a shot I guess as tedious as that sounds.

>>1135609
Yeah, I put all ground wires on 1 common ground rail (uC and driver) in my test. If I isolate my diver ground from the uC ground I just get the 4th case for all input LED arrangements (full voltage output, no pulsing).

File: download.jpg (4KB, 276x183px) Image search: [Google]

4KB, 276x183px

>>1135623
I have never seen an SMD resistor or SMD passive component that didn't look like pic related

can you take a pic of one out of the strip? preferably one that isn't blurry

What does /ohm/ think of Oyvind Dahl? Neat dude or terrible charlatan?

>>1135623

On the basis that they came with LEDs I'd assume resistors. Can you take a closer picture of one of them?

>>1135623
they look like side emitting LEDs

put a multimeter on diode and see if they light up dimly

>>1135648
Yeah, this.
Or at least post less potato pic.

I've got plenty of 18650s and want to make a power bank for cheap.
What's the cheapest way to step up 3v to 5v with relatively high efficiency? I don't mind if current is limited to 0.5A if it means 90% efficiency, but I wouldn't want any slower.

Off topic, but how old are you /ohm/? I'm curious about the kind of people who come here.

Bonus: what you do for a living

File: Untitled-1.jpg (2MB, 8308x1851px) Image search: [Google]

2MB, 8308x1851px

>>1135623
>>1135627
>>1135640
>>1135676
Here's the closest pics I could get, my lens doesn't do macro all that great.

>>1135648
I'm not sure if I even have a multimeter. I can check later tonight

>>1135815
Yeah, those look like LEDs

>>1135816
Maybe they sent me the wrong thing then and I just now noticed it. Thanks anon

File: 395ebe35979caebd81cb958a3b373e04_l[1].jpg (26KB, 360x360px) Image search: [Google]

26KB, 360x360px

>>1135815
Those are side emitting LEDs.

>>1135601
I'll buy it from element14 then.
Is there anything better for a similar price if I want to build a power bank to charge my phone?
http://au.element14.com/maxim-integrated-products/max1763eee/dc-dc-conv-sync-boost-1mhz-qsop/dp/2513942

What's the difference between a regular optocoupler and an opto TRIAC?

As far as i understand both are used to completely separate two circuits and make sure nothing gets mixed, but is there more to it?

File: 2dd.jpg (37KB, 600x600px) Image search: [Google]

37KB, 600x600px

>Get Make: Electronics book
>Experiment 1: Let's lick a battery!
>Experiment 2: Let's short a battery!

>>1135813
Im 18 and about to start electronic engineering in a good university in a third world country

>>1135813
22, meche, defense industry

whats a good adjustable power supply for prototyping

>>1135999
You take one of those transformers that go from 3.3 to 12v (or whatever you need), cut it's tail and solder two jumpers (the same ones you use in your proto board, of course) where the head used to be. That's the most cheap, basic setup, and if you need something more specific you'll just have to design one for that.

>>1135813
22 final year EE, wireless power design
tesla was a hack lamo

File: 0-60-volt-dc-variable-power-supply-using-lm317-lm337.jpg (320KB, 1224x770px) Image search: [Google]

320KB, 1224x770px

>>1135861
>What's the difference between a regular optocoupler and an opto TRIAC?

the regular ones have a phototransistor output for operating DC stuff, whereas the triac-output ones are for controlling AC stuff. generally the triacs in these things are so weak, you would actually use an external triac to control a load of any substance.
the most commonly used part for AC loads is the MOC3010 which has a diac which turns on a external triac.

>>1135999
you can make your own like in the pic, or get a Mastech HY1803D, or equivalent, if you need current limiting, which you do, coz you're a noob who's gonna burn your share of chips before you get wise.

>>1135075
Roll

>>1135813
19, second year physics student, I solder out of a briefcase at a university hostel.

File: BLDCControl.gif (39KB, 720x386px) Image search: [Google]

39KB, 720x386px

>>1135509
I think the problem is that you're assuming you can control this motor by flickering each terminal between 10V and 0V. If I remember correctly, to control these 3 phase BLDC motors, you'll want to flicker between +10V and -10V, which (I think) changes the direction of magnetic flux due to the changes in electrical current direction in said coils. Replace those PFETs with NFETs, then try pulsing the gate of each NFET individually. You'll need to make a feedback system that reads the backemf to properly time your pulses to each NFET (if it's sensorless).

Pic related, anyone want to confirm? I vaguely remember this from my abhorrent electromechanics course in 3rd year.

>>1136081
you're right. you can run them properly by connecting three comparators from the center tap to each of the three phases and catching the zero cross on the idle phase to determine when to progress the driver state. it's not strictly necessary just to get one to spin though.

the pfets are fine but he needs a separate gate driver/pin for them. an r-npn would work. his waveform is obviously wrong since you need tristate control of each leg which complimentary wiring won't do.

File: 1470008080844.png (129KB, 1291x884px) Image search: [Google]

129KB, 1291x884px

>tfw you're not sure if you wired anything wrong but you're too lazy to check

File: CDROM stepper.gif (2MB, 480x640px) Image search: [Google]

2MB, 480x640px

Got this to work with simple pulses.
Cutoff in pic is +5Vdc

File: IMG_20170226_070055_EDIT_1.jpg (1MB, 1976x3424px) Image search: [Google]

1MB, 1976x3424px

Does the solar panel on this thing do shit or is it just decorative?

>>1136208
drain the battery and find out you dunce.

>>1136208
They're usually pretty feeble, especially for a multi-amp-hour battery. A big problem with the circuit used with them is that the LEDs light up whenever there's power coming into the panel, so it will waste power though the LEDs if there's hardly any light. Make sure to cover the panel when you're not using it if it does have this issue.

>>1133933
Finally got the stupid thing to light up

>>1136211
Just noticed the LED lights up but only in direct sunlight.

Would this be good enough if I'm going camping?

File: IMG_20170226_072457.jpg (3MB, 4160x3120px) Image search: [Google]

3MB, 4160x3120px

>>1136209
I'm trying but my phone is already at 100%

I guess it's nice that it can charge my phone by at least 60% and still have leftover...

Questions cool here? I havent been here in forever.

How would I go about connecting 2 keyboards to a PC but rebinding keys on the second keyboard so that it won't affect the primary keyboard? Basically bind unique keycodes or macros.

I am aware there are some programmable keypads that do this, but I would like to repurpose a numpad.

>>1136222
Only in direct sunlight? Should be fine then. Watch out though, the mAh rating might not be accurate. An iPhone 7 Plus has about 3000 mAh, so ideally dividing your battery pack's mAh value by 2000-3000 will get you how many times it will charge your phone. In reality, there are losses of up to 20% or so, but your pack should have a few times more capacity than a phone. See this video on misleading power packs:
https://youtu.be/NL6nySngGtU

>>1136222
For camping you should be able to strap that thing to the top of your pack or leave it out in the sun for a few hours, just check it to see if it gets hot. Bonus points for waterproofing.

If I want to make an rfid reader stronger with an amplifier circuit, am I right in thinking that I don't need to connect the reader to the output? Surely so long as I tune the antenna's power supply to the correct frequency and be careful to avoid noise, then I don't need an amplifier circuit?

>>1136081
>>1136085
That would explain my motor jittering since I'm only energizing the coils in 1 direction.

I removed the connection to each gate and made a new code so that it spits out 6 signals instead of 3. However, now my motor won't spin at all. Taking a look at my output it appears everything just permanently stays on.

>>1136208
I have one of these
Left it in the window for a few days and it didnt charge the battery much or at all.
I opened it up and checked to see if the solar panel even did anything, desoldered the battery leads off the board inside and put the solar panel in the sun and it was outputting a charging voltage. I dunno mang, it just seems that the solar panel on it is so weak it might as well not be there.

>>1135075
Rolling.

>>1135813
23
chef
electronics hobbyist though. Moslty I'm here to fuck around with turning LEDs on and off

>>1136248
>waterproofing
It's got rubber seals for the ports, but since it's a cheap Chinese one I'd rather go with water resist than waterproof.

I left it in the sun for 7 hours and battery went up by 1 bar on it so I guess solar isn't that potent...

>>1134360

Whats the drain to source voltage on your PMOS transistors? Are you switching them from the linear region to cutoff?

If you aren't switching from the PMOS ON and the NMOS OFF to PMOS off/NMOS on then you will have periods in which both the NMOS and the PMOS are on. When that happens, you will short out the supply through the NMOS and PMOS transistors.

You need to make sure that the high voltage on your PWM signal is high enough to force the PMOS transistor to turn off. From all of the diagrams shown so far - I would expect your PMOS to stay on, all the time.

Here's an experiment you can do to find out if this is s problem: Set the supply voltage drive the motor to (12Vs) to 5Vs on your original circuit. Replace the motor connections with three small resistive loads to ground (Like 100 Ohms). Now with the PWM set at 50%, play with the supply voltage and observe the behavior of your circuit. When the circuit is behaving correctly, the supply should give ~3*(Vsupply/Rload * 0.5 + 0*0.5) amps average. On the logic highs, the NMOS transistor turns on and shorts the load to ground, while the PMOS transistor (should) turn off. On logic lows, the PMOS transistor turns on and shorts the load to the supply while the NMOS transistor turns off. If you increase the duty cycle, you should find that your circuit consumes less power - and vice versa. I suspect your circuit should have a (for this resistive load to ground case) current of ~3 * (1 - D) * Supply/Rload (with D == Duty Cycle).

The problem is: If your PMOS doesn't turn off completely when the NMOS turns on, then you will see spikes of current that will bring the average up because the NMOS channel resistance is much lower than 100 ohms.

If you don't have an oscilloscope, I recommend downloading ltspice (its free) and using their pre-built components to make a mockup of your circuit. You can probe the currents through the transistors and confirm if they are acting in the undesired feedthrough manner.

>>1136420 - please see:

>>1136451

File: 1486970805303.png (66KB, 400x250px) Image search: [Google]

66KB, 400x250px

hello /diy/
this first week here and im starting collage for electronic engneer.
im going to read the getting started in electronics by forest mims II
what projects should i take on for a beginner like me? any tips?

>>1136471
get an arduino and make a clock with a 7 segment display or a random number generator

hard mode: do it without an arduino

>>1136471
Which college? Just out of curiosity

File: X4715A0.jpg (728KB, 1920x1200px) Image search: [Google]

728KB, 1920x1200px

I want to put an outdoor camera, powered by a battery, and controlled by a single board w/ micro sd, what brand/camera do you recommend? knowing they could be damaged by rain

>>1136558
what do you want this camera to do?

>>1136560
baking pancakes

File: 9TxZtVz.jpg (527KB, 1899x1159px) Image search: [Google]

527KB, 1899x1159px

>>1136560
>>1136561

take a picture every 10s till the battery is dead, or the sdcard full,

pancakes would be a plus :)

>>1136471
what do you know already about electricity/electronics

File: 1484531829536.png (46KB, 520x673px) Image search: [Google]

46KB, 520x673px

>>1136420
i drew you a monstrosity. the npn can be a 3904. all of the resistances should be roughly as small as you can make them while respecting dissipation and max pin currents.

>>1133969
what about reading some fucking datasheet you retard

Copy-pasting from /sqt/

How hard would it be to make the touchpad controller from this game?
https://www.youtube.com/watch?v=yYs4vMcxKO4

I played the game a couple (read: a lot of) times when I was in Japan, but it wasn't clear to me whether the input was discrete (effectively buttons) or if the buttons changing color was just a visual effect and the fingers touch points were tracked continuously. This doesn't matter for most of the inputs but there are notes where you have to a slide input to hit them.

Also not seen in the video are motion sensors above the touchpad to detect the player raising their hands for the air notes. I'm not really interested in duplicating those.

>>1136507
community college

>>1136567
im on the first chapter of the book
and i understand small basic stuff of how eletronics works
the only project i ever did was connecting clips from a LED light to a 9 volt battery
i also know how to do soldering

>>1136709
Don't go into EE. Just do Computer Science or Software engineering.

EE is a mature industry so you will need a masters to do anything "design" related and a PhD to do interesting designs.

I'm on a PhD track and I wish I pursued underwater basket weaving. FML.

>>1136721
do you have experience with CS that makes you believe the experience for 95% of people in that field won't be the same as in a mature industry?

>>1136721
well im planning to get a PhD anyways for EE

>>1136447
W-well at least it's probably got high battery capacity

>>1136725

In my experience, there seems to be two factors that make CS better for the individual. First, the demand for people who write code is higher for those who design hardware. As long as the demand is higher, the expectations for the individual engineer are lower. Second, hardware engineers have high non-labor specific NRE and RE costs. So you want only really good engineers who don't "create" problems post fabrication through poor design. Software engineers have relatively small material costs and software, in the modern era, is mostly patchable.

Another problem with the EE field, due to the low demand, is the really great high paying jobs are hoarded by top tier schools. I went to Stanford for grad school. If I wanted to work at Apple for a summer ( as an intern making 40+ an hour), then all I had to do was be a warm body at the career fair in front of the apple booth. At my undergrad, it took considerable effort to get a test engineering intern position for 23 a hour. The full time offer I was given afterward? 35 ish an hour. They paid me less than I would get paid if I interned at Apple.

CS doesn't have as much of a problem due to the huge demand. Microsoft came to both universities and hired people for the same rates.

>>1136562
Trail cams are a thing, my dude.

>>1136721
>EE is a mature industry so you will need a masters to do anything "design" related and a PhD to do interesting designs.
True.

>I'm on a PhD track and I wish I pursued underwater basket weaving. FML.
Dunno about that man.

>>1136862
Ahh - I mean research is difficult and there is not always an incremental step forward. Its easy to get bogged down in a swamp of work and not be able to see a path outward.

>>1133933
>>1136218
I made it into an oscilloscope over the weekend, can share details if anyone is interested.
Pretty neat so far, and a lot more I can do to improve it

what kind of flux should I get for manual soldering (ie not reflow)?

>>1136925
Standard is no-clean flux. If you get water-soluble flux it needs to be washed off (it's acidic or something).

>>1136721
>EE is a mature industry so you will need a masters to do anything "design" related and a PhD to do interesting designs.

Seriously? I was told by profs/students alike that a master's is useless so I didn't bother.

Do you mean to tell me I have to go back to school now if I want to do PCB/FPGA design professionally?

>>1136969
The way it works in Australia is that you whore yourself out during your undergrad to get into a postgrad program at a big firm if you want a career and you do postgraduate studies if you want to stay in academia forever or eventually say fuck it and do a trade instead.

>>1136973
I'm a fucking leaf and as far as I've seen in terms of jobs here and in the states; there isn't much available for master's, everyone just wants bachelor's with work experience. Companies that have turned me down said the best way to stand out to them and improve my chances of getting hired is by doing shit outside of studies (aka personal projects) because they prefer practical experience that fresh grads tend to not have.

Hence why I'm even on this general, trying to build my portfolio.

I don't want to go back to school just to be specialized in 1 facet of EE. Like yeah, I want to do PCB/FPGA stuff now. But I'd rather be flexible so that say 5 years down the line I want to do transmission lines or someshit, I can switch over. I imagine getting a masters would limit you to a particular industry.

>>1136969
>>1136976
What do you do now? The big companies don't like hiring you to do something you haven't already done, which is kind of a catch-22.

If you have impressive enough side projects it might help but generally work/school projects are weighted more heavily.

>>1136986
The problem is that school projects are just literally the same crap every other jackass applying for the same position has. That's why they don't hire you for something you haven't already done; you just completed degree requirements as far as they're concerned. You need to somehow try stand out from every chang, thad and patel who have the same base knowledge as you and the same degree. Work experience will get you noticed and push you to the top of their stacks, but as a fresh grad looking to get a job, it is a catch-22.

What would even be some "impressive" side projects to beef up a portfolio? Coming from someone that mainly wants to do PCB design.

>>1136996
Design your own FPGA board, maybe an SDR or something.

Stuff I've thought about doing: GPS receiver from scratch, 802.11 client, graphics card, wider bandwidth N2PK VNA.

File: Triadtron%20Inverted%20Amplifier%208[1].png (66KB, 570x491px) Image search: [Google]

66KB, 570x491px

the absolute madman

File: L001j.png (11KB, 262x272px) Image search: [Google]

11KB, 262x272px

I'don't know much about transistors, and I'm trying to build an RS latch with two PNP transistors, basically pic related but in reverse
It's latching on as I want, I'm getting 5 Volts on one output, but the other is at 2 Volts and I would like that to be lower. do I have to change the resistors going across? or can I do something with the outputs to bring them to a level I can use (0/5 V)?

>>1137207
I'm not too sure where are you referring your 2V, but you could try bypassing your LEDs with suitable resistors so that the LED voltage drop wouldn't lower the output voltage swing.

>>1137207
Literally just mirror the entire schematic vertically and then flip the transistors and LEDs around so the arrows point the correct direction. Vcc -> pnp emitter -> 470R -> LED -> gnd. The collectors and bases are cross-coupled with 10k resistors, then add the switches between the bases and Vcc.

>>1137221
yeah it was the leds, removing them fixed it
>>1137270
that's what I meant with in reverse

I use the output of the latch to drive an lcd, but it seems the pixels on the lcd appear dimmer, I'm not sure why though, maybe not enough current? I tried using the latch to control a third transistor, (wired like in the pic, where it says short to ground is connected to one of the collectors in the latch), but I'm not seeing a lot of improvement. what could I do? Is it even possible to get the same performance as connecting the circuit to VCC directly?

File: download.png (3KB, 210x212px) Image search: [Google]

3KB, 210x212px

>>1137311
forgot pic
it seems lowering R1 increases the brightness, but not all the way, shorting it doesn't work either

>315mhz rx picks up noise when i play a youtube video
w h y

>>1136916
A DIY oscilloscope for cheap? Count me in!

yo h/ohm/ies

I don't have a soldering iron holder. If I put my soldering iron into an empty beer bottle, is it going to melt/explode the bottle from the heat?

File: Throwies.jpg (60KB, 1064x798px) Image search: [Google]

60KB, 1064x798px

Alright /ohm/, I've got a simple question with a probably-not-so-simple answer:

tl;dr is I want to make a long-lasting throwie. How?

Longer version: I want to make little "beacons" which I can stick around my house which will last for a while. It gets dark and my light switches are far apart and not near where I need them. I can make throwies, but AFAIK those are good for maybe a few days at a time and super inefficient. What do I need to do to the circuit to make it longer-lasting (on the order of weeks, or months if possible)?

Another issue is that I need the footprint to be as small as possible. Regular coin cells (20mm) are way too big, and I know nothing about any batteries other than everyday use ones such as AA etc.

Suggestions?

>>1137546

don't be autistic and buy those night lights that plug into the outlet and turn on when it's dark

File: IMG_1833.jpg (83KB, 625x313px) Image search: [Google]

83KB, 625x313px

>>1137546
You can consider hooking it directly to the wires inside the switch. Maybe stick an inverter or something behind the face plate? Be very careful though and take all necessary precautions. People die by getting electrified every day and it's extreamly sad for their families, but don't let that distract you from the fact that in 1998 The Undertaker threw Mankind off Hell In A Cell, and plummeted 16 ft through an announcer's table.

>>1137546
> It gets dark and my light switches are far apart and not near where I need them.
Put the LEDs or preferably neon bulbs inside the light switches, parallel to the switch contacts. The drawback is that you need to tinker with the mains power, so consider >>1137554 instead.

To answer your actual question:larger batteries, less current to the LEDs (and higher efficiency LEDs), blink them at low duty cycle to minimize current consumption. Add a twilight switch to them, too.

>>1137555
>>1137555
Hold on, I'm not wanting magnets to put inside the switches nor for my switches to light up - the point is for me to be able to put these where I need, e.g. bookshelf, table, bedside, etc. and not just an electrical switch or plug.

>>1137557
>larger batteries
Dang. Obvious, but still not desirable.

>Less current to the LEDs
The two people I've talked to said resistors would work but the energy would be given off in heat and not make the things last any longer. Don't know if this is is true.

>blink them
Only way I know how to blink things is a 555 + some more ICs, but that's too big for this if it's anything like I remember HS digital logic circuits were.

>twilight switch
I assume this turns it off during the day? Still runs into the problem of taking up space, plus I'd like to make it aesthetic and not just circuits slapped together in a pile.

I'm not outright rejecting the ideas, just elaborating on my needs/wants - I might go a few sizes larger based on what I'm seeing so far but ideally I want it as small as possible.

>>1133969
Yo man share your findings i'm interested in wearables too

>>1137560
>said resistors would work but the energy would be given off in heat and not make the things last any longer.
Resistors are the simplest, smallest, cheapest way to reduce the LED current. Also, when you drop the current to microamp level, beating the resistors in efficiency becomes increasingly difficult and also pretty irrelevant. One button cell lasts something like year when you draw only 10µA from it.
The obvious drawback is that the lower the LED current is, the dimmer it is. Even the brightest LEDs are pretty dim at 10µA. Might be enough in darkness, though.

>Only way I know how to blink things is a 555 + some more ICs, but that's too big
Use SMD components or some dedicated LED blinker ICs. One such IC is HK3258, but it is
pointlessly power-hungry.

>I'm not outright rejecting the ideas
You pretty much are. To put it differently, most of the time physics works against you and when it doesn't, your lack of equipment, skills and knowledge seems to do the job.

>>1137560
Low-power microcontroller with sleep mode. BAM

>>1136976
>>1136969

When I mean design, I mean IC design or System design. If you are taking EE courses to work at an engineering firm that builds breakout or development boards then you will trade job complexity for significantly lower pay.

I make $65/hr US dollars as a Research Engineer/COOP at a major US company.

However, my boss doesn't come to me and say "We need to build a board that this FPGA in this form factor." Instead, he comes to me and says, "How can we design an IC that minimizes the energy consumed for a particular computation." OR "What is the feasibility of using the work done in this paper to advance our process/system?"

The end result is that I either give him a specification that then a team of engineers could implement with some justification or give him practical limitations why such a thing can't be done (although the overly academic engineers with PhD will forget reality and forget to remember what is practical...)

If you want to make truly transformational amount of money, then you want to be the person who designs the system, not the person in the team who implements it.

Of course, there is a lot of stress and board design can be quite relaxing. I made a board that links a Tegra TX1 and an FPGA/SoC for fun basically with a friend.

>>1137546
[spoiler] use a joule thief [/spoiler]

>>1137534
What do you want to know about it? I can get pictures later. I pretty much took a TI mcu and that oled screen I posted above, read inputs through the adc, then do some graphics manipulation to plot points in the right place. Simple once I got the screen to actually light up.

>>1137581
I'm confused by your post. So you're suggesting I stay away from PCB design because it pays shit?

I am a fresh grad so I'm looking to take anyone that will hire me.

File: 1454350828379.jpg (68KB, 668x712px) Image search: [Google]

68KB, 668x712px

>interview for assembly/electronics job
>only got into soldering recently
>right anon can you solder this please
>fuck
>solder it, shit at first then get used to it
>'That's hoinestly the best soldering for a new person I've ever seen'
>later on in the interview they get me to solder again to show another bloke
>do absolutely shit apart from the final pcb

Woman said I was near the top of list in terms of getting the job but fucking hell I cocked up.

I don't have the money for a decent power supply, whats the next best thing i can get? I only need the standard 3v 5v 12v, since thats enough for arduino

>>1137828
Wall wart + 78xx IC?

>https://en.wikipedia.org/wiki/78xx

>>1137828
Go dumpster diving for a computer power supply.

>>1137828
Power supply from an old computer for 5V(red wire) & 12V(yellow wire)
3V regulator for 3V

>>1137837
3.3V is also available via orange wire btw.

>>1137835
>>1137837
thanks for the suggestion, already found one local for 8 euro

>>1137851
It's actually a meme but ok

>>1137747
how the f*ck can you poorfriends hold such a demeaning job without killing yourselves

when i think of a job soldering pcbs i think of a 5 year old chinese boy that's gone blind in one eye and has a cataract in the other

>>1137747

I once got hired on the spot in an interview when they asked me to use a MIG welder. I was applying for an electromechanical assembly job and the job ad was asking for people that could solder and read schematics. I never used a MIG welder before but they said it was fine. I went to weld a cap on a thermocouple and accidentally grounded myself on the grounding block and got a shock. I shook my hand and went back to trying to weld the cap. The guy stopped me and told me I was hired. Most applicants quit after they got shocked. I never had to use the MIG welder after I got hired....they had full time welders for that.

>>1135813
18, first year EE student. As of now I sell Adderall to brainlets and work at the law library.

>>1137913
I laff'd

Anybody know what this thing is?
I'm taking apart stuff around my house as a learning experience, this is a solar powered lamp. I can recognise the resistors, I assume the tall black things are some kind of transistors to switch between the rechargable battery and the solar panel.. not sure what this little black thing is though.

File: Lamp.png (637KB, 1096x381px) Image search: [Google]

637KB, 1096x381px

>>1137963
forgot pic

>>1137967
that's a diode. it allows current to flow in mostly one direction. i assume it's for reverse polarity protection. how nice of them to sacrifice 10% efficiency for your safety.

>>1137945
>Most applicants quit after they got shocked.
The rest quit after their pacemaker fucked out and they died, right?

I do phone screens and interviews for new EE grads pretty regularly. I'll echo what others have said - people with personal projects get moved to the top of the stack.

>>1137982
is it fun? i always wanted to sit on interviews in my office but nobody wants designers' inputs on whether an applicant would be good at design.

If I have a chip "SN74LS393n" and I look at the pinnout/datasheet for "SN74LS393" will I get fucked

File: 1485056608619.png (82KB, 545x803px) Image search: [Google]

82KB, 545x803px

>>1137987
N is the package designator. it has a pinout specified in the datasheet.

>>1137988

yeah... I realized that 2 seconds after I posed it, shoulda RTFM

What is a good connector/wall wart to use as the 5v power in for all my perfboard projects?

>>1137967
A solar cell can consume electricity when connected to a battery but not receiving enough light to charge the battery.
The diode is a one-way valve that lets current flow from the cell into the battery but doesn't allow the battery to flow back into the cell when it's not producing.

File: 1470921835134.jpg (656KB, 768x1024px) Image search: [Google]

656KB, 768x1024px

ok guys
how do I make an fm transmitter with a range of about 10km from scratch with off-the-shelf basic components (instead of just buying one)
I got real basic skills but I can read a diagram, I just can't quite find what I'm looking for by myself and the concept of diy radio is neither legal or popular in my country.

>>1137991
whatever your old cell phone came with

>>1138011

Can't do microUSB on a perfboard without a painful bodge

>>1137585
This.

>>1138020
cut the connector off and wire directly

>>1138020
>a painful bodge
are you sure you want to be working with a perfbord little lord fauntleroy?

>>1138029

then I have to find a different wall wart for every little perfboard project

>>1137991
>>1138035
i took a 12v barrel plug wall wart and beheaded it and soldered two male header pins on it instead. you could probably do the same thing to a usb cable.

>>1138035
why

>>1138034

> surface mount soldering a MICRO usb port on a board spaced for DIP packages

>>1138038

because the wire is soldered into the board

>>1138037

> barrel plug

looks good

I will buy some on ebay

File: solar_schematic.png (18KB, 1920x1080px) Image search: [Google]

18KB, 1920x1080px

>>1137967
You flipped the image, I wasn't expecting that. But I got your schematic for you, more or less. Excuse the DaveCad.

So the left transistor turns on when it receives enough of a voltage across the battery as measured by the voltage divider (the 47kΩ and 100kΩ resistors, read about voltage dividers here: >>1135071 ), and when switched on, the transistor shorts the solar cell through the 10Ω power resistor as if to stop the battery from being overcharged. Here the diode must act as a sufficient enough voltage drop to stop the battery from charging, which is a little sketchy, though it should be a big enough diode. Though maybe photovoltaic cells act like current sources with a max voltage, who knows. In the other side of the circuit, the second transistor stops the LED from lighting if there isn't enough voltage to light it "enough", as triggered through the 3kΩ resistor. This isn't a strictly necessary feature, and it does take a little control away from the user, but it prevents the LED from preventing the battery from charging quickly by drawing a current and producing insignificant/unnecessary light.

>>1138084
Oh and the diode isn't in case you put the battery in backwards (reverse polarity protection), it's in series with the photovoltaic cell so that when the cell is shorted across the 10Ω, the battery isn't also shorted.

>>1138084
>>1138086
Yeah I figured that it'd be easier to follow and forget to say anything about it,
Thanks so much though dude that explains so much, didn't expect anybody to take much interest in my post. is it generally necessary to have all these measures in place to create a solar charged battery? What even happens if you over charge a battery?

>>1138121
It entirely depends on the battery, but put simply, they explode. A battery is a combination of two chemical half-reactions, one of which produces electrons, the other of which absorbs them. They absorb/produce ions of the opposite charge to balance the charges. A battery only has so much room for reactants and products, so when you put too high a voltage across it, the battery charges up to it's maximum, starts shorting internally, heats up, and expands. Your device probably has a couple of NiCd/NiMH cells in it, which pop. If it's a high end one, it will have Li-ion cells, which burst into flames. Consequently, Li-ion and LiFePO cells have to be charged with a carefully designed and regulated circuit, often with temperature feedback. If undercharged, they have to be charged at a constant current, before charged by constant voltage, and I don't think there are many domestic chargers that do this. But with a little know-how and a bench top power supply, you can do this yourself. A bench top power supply with constant current costs well upwards of 100 freedom bucks, while a Li-ion charging circuit can be bought from chinkistan for 36 cents, so don't over-discharge them.

At least now-days, circuits can tell when a battery is charged by the voltage. A car battery is below 10 when undercharged and above 13 when full, so just by putting that into an ADC, the control chip can trigger a FET and stop the charging. The nice thing about voltage-based charging regulation is that you can put as many cells in parallel as you want and nothing changes, but the standard BJT transistors in your circuit are current triggered, hence the resistors in place to limit the current. If you put in a smaller electrochemical cell or smaller photovoltaic cell, the maximum current sent through the respective transistor might not be enough to trigger it, stopping your circuit from working. This might also happen as your battery ages and loses capacity. Cont.

File: HTB1.IQmJXXXXXazXpXXq6xXFXXXm.jpg (110KB, 798x547px) Image search: [Google]

110KB, 798x547px

>>1136558
I got one of these 4k action cams, $45 or something, it's been running since December 27th (Plugged in) and hasn't stopped so far. 64GB card's still more than half full.
I forget what it's set to, I think a photo every 10 seconds.

>>1138146
You're not in any danger of overcharging the battery by covering the solar cell so that the overcharge transistor won't trigger, because when you cover part of the cell you'll be decreasing it's voltage, stopping it from charging the battery in the first place. If it's dim out, the current provided probably won't charge the battery anyway, so I guess it doesn't matter.

Now I certainly haven't seen any circuits like yours and couldn't tell you if it's normal or not, but this battery circuitry is pretty primitive. The closest thing I've seen would be my cheap USB solar power bank. With LED built-in. Not a lone transistor in sight, at least not a BJT. All the circuitry is on a big chip with writing on it, but I do remember the solar panel was pretty garbage, especially to charge five 18650 (big) cells. Typically, at least for USB and Li-ion applications, these circuits have some manner of voltage converter in them, converting the varying voltage of the battery and perhaps the solar cell into a constant 5V for USB or 3.7V to charge the batteries. In your application there is none of that, so I guess the battery charge (at least across the voltages where the second transistor is triggered) doesn't change enough to affect the LED brightness, and assuming there aren't dropper resistors inside the lamp itself, happens to be a good voltage for an LED. A normal NiMH voltage is about 1.2V, so if it's white there should be three battery cells. If there are four then there will be dropper resistors. It may also be a single Li-ion cell, the low voltage power-off could be an indicator of this.

What kind of lamp is this? Because it does have what might be a simple "auto-power-off" function, but it doesn't have anything stopping the light from coming on when there's light on the panel. Are the light and panel in seperate locations?

File: basic-garden-solar-light-circuit.jpg (53KB, 530x274px) Image search: [Google]

53KB, 530x274px

>>1138150
Ok, I might have made a slight blunder. Assuming that pic related and your circuit came from those garden solar lights, the right-most transistor will be a PNP, not a NPN, i.e. it requires a negative voltage at the base (the wire perpendicular to the centre line). This doesn't make sense to me because there isn't a voltage at all when the solar panel isn't receiving light, which is when the light is supposed to turn on. Unless PNPs trigger on 0V, which I doubt. Pic related does show a potential current path of from the battery +, into the emitter (why is it called emitter if it absorbs current), out the base, though the resistor, through the PV cell, and into the battery negative, hence potentially triggering the transistor, but your circuit shows none of this. Assuming I didn't mess that up too. I did bumblefuck my way through the circuit diagram and completely guess the BJT polarities and orientations after all.

Hi. First post here on /ohm.

how would you do to make a phantom power supply from 9V battery??

What power converter would you use?

File: led-1.jpg (164KB, 763x896px) Image search: [Google]

164KB, 763x896px

need help in identifying electric motor for wind generator conversion.
is this permanent magnet motor?

File: 4L_yubDOIkp.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

Just made this for fun.
It's a game that uses a 74164 shift register with a data loop. You basically press the PS button to try and stop the led in the middle and the CLR button to stop. The LEDs on the right display if you got it, were too early or was too late.

>>1133263
I just repaired my Good ol' ProView cold cathode lcd (replaced the caps), am I /yourguy/?

>>1133310
If you want a good enclosure you can just go to a metal shop they can cut stuff for you, it costed me 30$ for a decent looking power supply case.

>>1133525
Get some perma-board and make yourself an adjustable signal generator using a 555 and a pot.

>>1133525
Get yourself a logic kit (just a package with a bunch of different logic ICs)
I'll also assume you have a breadboard

>>1138193
>220v rating
>3 input terminals
>3000 RPM speed
It's a 50Hz, two-pole, three-phase motor. Not suitable as a generator.

>>1137913
Have to start with the shit jobs m8

Not smart enough to get three As at A level and do electronic engineering, I was always shit at maths.

>>1137913
I work as a avionics technician, a year out of college I was making 60k a year

File: tmp_10400-Screenshot_2017-03-01-19-44-13-11346414731.png (127KB, 1368x2023px) Image search: [Google]

127KB, 1368x2023px

I'm making a motor with a couple of servos, the battery voltage is 8.2V and the servos run at 7.2 volts max, accoding to chinese seller.
The voltage drop seems too small to use a voltage regulator, so I thought about using a pair of diodes and another diode to return voltage spikes to vcc. Also a resistor to add a small load to keep the voltage down.
Is there a better way to do this?
Ideally I would like the voltage to be at 7 volts all the time.
Is the red diode necessary?

Excuse the poor drawimg, I'm on the bus.

>>1138317
Fuck, the resistor should be on top of the motor.

File: 4L_yhqQWbAp.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1138317
The Zener is 7.2 and resist to ur hearts content

>>1138340
>>1138317
Also make sure to read the data sheet, it might be better to use a 5v Zener. Also the resistor would just lower current it would not affect the voltage.

>>1138340
That circuit won't work, it's too inefficient. The servo uses around 100mA, this means the resistor would waste 0.1 Watts in form of heat, also the servo isn't always on, so that circuit will be wasting battery all the time.
100mA times 7 volts would be an extra 0.7 watts wasted, for a whooping 0.8 watts wasted x2 servos = 1.6W wasted all the time.

>>1138317
>7.2 volts max
There's a good chance this actually means that the servo is intended to be used with a 6 cell NiMH battery, which has a nominal output voltage of 7.2V. The maximum voltage of such battery (when fully charged) is over 9V.

>>1138365
-where did you learn your math
-yes it would work I've used that exact schematic before.
-use a regulator if you want high current
-you can't have high current without high power loss
You fucking idiot

>>1138193
Looks like a stepper motor to me, those things are quite useful in CNC, assuming that that's what it is.

>>1138243
>Drawing schematics with a ruler

Your autism is showing

>>1138418
It's a shame I also get paid 60k a year for drawing schematics with a ruler.

I assume you use the "draw shitty unorganized lines like a tard" method?

>>1138420

He doesn't 'do' anything.
Eating Cheetos and ridiculing others who are productive from the comfort of your basement 'battle station' isn't doing anything.

>>1138420
>he takes offense to being called an autist

I was actually complementing you you fucking self righteous turbo faggot.

>>>/reddit/

>>1138432
My apologies for taking that as an insult, I think you're an autist as well.

>>>ifunny.com

>>1138434
Apology accepted :)

>>1138411
>where did you learn your math
First, that circuit requires a current limiter resistor or else the zener will fry, that resistor has to let enough current pass to power the motor, in this case 100mA plus the current to make the zener work. 8.4 volts minus 7.2 volts equals 1.2 volts, in order to achieve 100 mA with that voltage drop a 10 ohm resistor is needed, ending with a usable current of 120mA. 120mA^2 times 10 ohm equals the power, which is 0.14 Watts.
Power equals current times voltage drop, thus 7.2 volts times 120mA equals 0.86 watts, the power of the resistor + the power of the zener is equal to 1 Watt, I have two servos, so that would be 2 Watts when the servos aren't even working, since the zener has no other way to limit the current other than the resistor on top, and when the servo is off that current needs to go somewhere, so it drains through the zener.
Second, motors are noisy, and they can demand current in spikes. The zener could stop regulating the voltage if the current is not enough for the zener to operate.

>yes it would work I've used that exact schematic before.
It can work, but it will probably fry the electronics in my servo, also I would need high power rated components, which is stupid, since I could achieve the same thing with 2 fucking 4007 diodes.
Also, those 2 Watts will be a fucking problem when it comes to battery life.

>use a regulator if you want high current
>>1138317
>The voltage drop seems too small to use a voltage regulator

>you can't have high current without high power loss
It's called a buck converter. Using the two diodes I would lose 0.24 watts, not 2 fucking watts all the fucking time like your stupid circuit.

>You fucking idiot
Says the guy who provides a circuit an order of magnitude less efficient than my already crappy design and gets offended when somebody points it out.

Fuck you.

>>1138371
I'll read the datasheets for the IC's on the servo's control board, but the motor could fry too.

>>1138420
>60k

Poorfag confirmed.

fac: the median income on 4chan is 300k starting any job you want

>>1138496
fact: you are butthurt

>>1138496
fact: you are technician for a reason and showed you don't know the fundamentals of electrical engineering

File: 1472781021313.png (6KB, 195x259px) Image search: [Google]

6KB, 195x259px

>>1138499
i'm not him, i'm a brainlord engineer

i wouldn't mind doing blue collar work though the mechanics at my company are unionized and make bank

>>1138501
Hopefully they save some of that money when their bodies begin to break down.

>>1135075
rollin

>>1138472
if your servos are not going to be activated all the time why the fuck are you using a uncontrollable circuit. if your so intelligent go build a goddamn digitally controlled circuit, also i did the math and the current limiting resistor (which is clearly in the drawing) would be 14.4 ohm.

throwing a resistor in parallel with your circuit isnt magically going to make it stay at 100mA i fucking dare you to build what you just drew up and keep it under your bed for the night.

>high power rated components
where the fuck are you getting that from? your powering it with 8 volts you dont need anything more then hobby level components.

do you understand how buck converters work? they trade V for I I.e. 2V 1A 3W-----> 1V 2A 3W
the power is always the same but i guess you dont understand because you cant even identify the most basic zener power supply out there.

dont bitch that your circuit is inefficient, im sorry that my circuit has a higher wattage but atleast it woth burn your house down. if you draw high current you will burn a shit ton of power it doesn't matter if your fucking god P=V*I High I ========= High power. unless your using 0.0000000001 volts your gonna have to grab your balls and live with the fact that youll loose power.

>>1138493
Considering im 2 years out of high school have 4k in my bank account and have 0 debt im actually in the 30%. also 50k to 60k in two years sounds pretty good to me.

>>1138523
ignore them, 4chan small board culture is elitism and condescension. don't brag about 4k in the bank though lol.

What's the minimum clock on most 74 series logic

will it work if I just use a button as the clock until I debug my circuit?

>>1138528
most 74s have no minimum clock. some have minimum slews on the order of 100us.

>>1138528
there is no minimum clock, but i would suggest De-bouncing your button, eevblog did a really good video about this. https://www.youtube.com/watch?v=Nj-Q8FQxHhU

Got word from a friend that works at radioshack that they are about to go under again. Every store will be having "everything has to go' sales. and it will all be heavily discounted. look out for this over the next month at your local Radioshack stores!

>>1138544
>feels bad to live in Australasia
When will someone tell me when the Jaycar garage sale is going on? EEVman went to a one in Australia, but no news for NZ.

>>1138544
i thought radioshack wasbalready extinct since about 2014? there are still stores left?

>>1138547
Ikr.

There aren't even any online sales here. Maxim 1763 is $15 here, it's cheaper to buy a Xiaomi power bank since it comes with lots of other goodies too like 3100mAh 18650s and other electronic parts.

>>1138548
They publicly went bankrupt but were saved after a buyout. They have kept a good amount of the stores open. but have had little sales because everyone thinks they all closed. This time they are going out for good. I guess its time for me to buy some half priced breadboards.

>>1138523
>4k in my bank account
>bragging about it

t. technician

>>1138522
>can't admit that his design is inefficient
>lashes out like an autist
>shows his own retardation

CS and applied math major here. embedded device programming fascinates me, is this still a viable career in the 21st century?
Will I have to imigrate to China?

File: 20170302_134024.jpg (1MB, 2048x1536px) Image search: [Google]

1MB, 2048x1536px

>>1138522
You dense motherfucker
>if your servos are not going to be activated all the time why the fuck are you using a uncontrollable circuit
Servos have a signal pin for something, when they are not receiving any signal they act like an open circuit (except for the electronics which control them). You fail to realize that, why the fuck would I be using a servo activated all the time? I could use a reduction motor for that. Servos are meant to be controlled to move things around.

>where the fuck are you getting that from? your powering it with 8 volts you dont need anything more then hobby level components.
The diode you've mentioned has a maximum power dissipation equal to 500mW, voltage times current equals power, 7.2 volts dropped times 0.1 amps equals 720mW, you fuckwit.

>i fucking dare you to build what you just drew up and keep it under your bed for the night.
I would rather use my design (pic related) than yours, since the zener diodes will probably catch on fire and burn my house, while mine only dissipates around 0.24 watts, since it only has a voltage drop of 1.2 volts, meaning each circuit dissipates 120mW

>High I ========= High power
>you can't have high current without high power loss
>do you understand how buck converters work? they trade V for I I.e. 2V 1A 3W-----> 1V 2A 3W

Are you a troll? I seriously have problems measuring the magnitude of your stupidity. You just described a 100% efficient buck converter without any power loss, which is what an ideal buck converter would do, when you stated in your previous post that I can't have high current without high power loss. Buck converters are not 100% efficient, but still, the power loss is small.

>but atleast it woth burn your house down
Unless I stop the servo, then the diode overheats and burns my house down.

>if you draw high current you will burn a shit ton of power it doesn't matter if your fucking god
But your circuit consumes 10 times as much power as mine when the servo is stopped

File: chain-shelf-hanging-chain-shelving-with-100-year-old-reclaimed-wood-10_fc488666-d2ff-4895-9afd-dca0e2b74764_grande.jpg (29KB, 450x600px) Image search: [Google]

29KB, 450x600px

Hello /diy/. I have a lot of books in my apartment. The floor there is also rather uneven (old building). The amount of books combined with floors fucked over existing bookshelves. They're literally tearing themselves apart, slowly.

For this reason I want to build chain bookshelves (picrelated). I can hang them from the apartment ceiling/walls. The question is: how do I mount them to walls/ceiling? They should be able to hold a fair amount of weight, like 200 pounds each, at least. I can't make mounts go all the way trough the ceiling, since, again, I live in an apartment. Any pointers?

posted it in /ohm/ because electronics are cool

>>1138150
>>1138153
>>1138146
It's a porch lamp but it's fucking awful. Not bright at all. It runs on a Li-ion battery, 400mAh's and shared across like 10? LEDs. I dunno how the fuck they invisioned this thing to work or even retain a charge.

File: 20170302_142423.jpg (2MB, 4128x2322px) Image search: [Google]

2MB, 4128x2322px

What is this called?
Are there suppliers?
Its a 8cm x 5.5cm laminated circuit with the ribbon cable coming out the back with 3M tape on it.
Its got 6 pushbuttons and a led on it.

>>1138774
hardware button debugger

>>1138774
It's for cyborgs to plug into and run code with. Humans augmented with a digital processor can think quickly enough to issue assembly instructions in real time and directly control a computer processor.

>>1138774

it's just called a "tactile dome button panel"

it was more than likely made for a product with a plastic overlay that said what the buttons did. It probably had been designed for that specific product only.

new thread

>>1138806