/ohm/ - Electronics general: hobbyist edition

[Anonymous No.998973]

previous thread: >>987082

http://pastebin.com/9UgLjyND

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

>What books are there?

Beginner:
Getting Started in Electronics Forrest Mims III
Make: Electronics Charles Platt
How to Diagnose and Fix Everything Electronic Michael Jay Greier

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

Advanced:
The Art of Electronics by Paul Horowitz

>What YouTube channels are there?
https://www.youtube.com/user/mjlorton
https://www.youtube.com/user/paceworldwide
https://www.youtube.com/user/eevblog
https://www.youtube.com/user/EcProjects
https://www.youtube.com/user/greatscottlab
https://www.youtube.com/user/mikeselectricstuff
https://www.youtube.com/user/AfroTechMods
https://www.youtube.com/user/Photonvids
https://www.youtube.com/user/sdgelectronics
https://www.youtube.com/user/TheSignalPathBlog

>What websites feature electronics projects or ideas?
http://adafruit.com
http://instructables.com/tag/type-id/category-technology/
http://makezine.com/category/electronics/

>Where do I get components and lab equipment from?
digikey.com
jameco.com
sparkfun.com
ramseyelectronics.com
allelectronics.com
futurlec.com
ladyada.net/library/procure/hobbyist.html
mouser.com
alliedelec.com
newark.com
ebay.com

>What circuit sim software do you use?
This mostly comes down to personal preference. These are the most common ones though:
NI Multisim
LTSpice
CircuitLab
iCircuit for Macs

>What software should I use to print circuits?
Circuit Wizard
ExpressPCB
EAGLE

[Anonymous No.999016]

Anyone else notice /ohm/ has been super sloe lately? Slower than usual.

[Anonymous No.999021]

>>998973
I gotta wind myself some RF coils.
Should I put the secondary winding next to or ontop of the primary winding?

[Anonymous No.999027]

>>999021
Bifilar winding is better, so the B. It removes self inductance quite well, and in transformers will remove magnetic flux.

If you wanna know more search about bifilar coils, I only just about got the grasp of how they're used.

[Anonymous No.999059]

>>999021
Next to, as that reduces the capacitance between windings. At RF, that can be significant.

[Anonymous No.999097]

>>999059
The primary problem with RF transformers tends to be the leakage inductance, not capacitance.

>>999027
He doesn't seem to mean bifilar, even though the pic looks like it.

>>999021
When bifilar winding (pic related) isn't possible, on top of each other produces lower leakage inductance (=higher coupling coefficient).
Is this more important than the low leakage capacitance depends on your requirements, but like I said, it tends to be, as isolation is usually less important in RF transformers than the power transfer.

[Anonymous No.999105]

>>999097
>doesn't seem to mean bifilar
By any chance, is this because of different no. of turns?
Isn't it possible to have both primary and secondary as parallel bifilar?

[Anonymous No.999114]

>>999027
>>999097
>>999105
>doesn't seem to mean bifilar
I didn't know about this.
I guess pic related describes what I originally ment better.
Would bifilar winding be an option for me if my no. of turns is different between the 2 windings?

[Anonymous No.999170]

>>999016
I'd say it's due to more exams for universities

[Anonymous No.999180]

>tfw studying computer science in Europe and forced to build an alarm clock
End my suffering, now. Americucks don't even know the real struggle of STEM. They even have to attend some literature and what not classes to fill up their credit points

[Anonymous No.999182]

does anyone know where to buy the metal boxes people use for their electronics projects?

[Anonymous No.999184]

>>999180
this is bait

[Anonymous No.999216]

>>999182
Welding.

[Anonymous No.999265]

>>999182
Aliexpress have lots of nice blank metal casings.
Or if you have knowledge in basic sheet metal you could just fold 2 pieces of galvanized iron and rivet them.

[Anonymous No.999270]

In software development, we call anything not defined by the creators of a programming language, "undefined behavior." Meaning that anyone that writes the compiler based on the programming language spec is allowed to do whatever they want, and anything could happen.

I know almost nothing about semiconductor circuits, but I would figure that "defined" behavior will occur when enough current can be sourced. What do you call "undefined" behavior, i.e. when a power supply can't provide enough current?

[Anonymous No.999272]

>>999270

i suppose the closest term in electronics is ''unknown state''.

[Anonymous No.999284]

>>999180
>computer science
>building alarm clock
wow you wrote some java or c++
fucking undergrads

[Anonymous No.999310]

>>999270
read up on brown out and brown out protection.

[Anonymous No.999325]

>>998973
Alright niggers: is a fluke 115 a damn good meter or nah? Worth the money? I don't want a piece of shit like I've dealt with up to now. Give me to cold, hard, raw, nasty, ragged truth.

[Anonymous No.999358]

>>999325
Anything Fluke is a damn good meter.

However, this is better and cheaper.

http://www.ebay.com/itm/FLUKE-15B-Digital-multimeter-Tester-DMM-with-TL75-test-leads-NEW-F15B-/331279222293?hash=item4d21c7c615:g:6yIAAOSwPe1T13Qp

[Anonymous No.999367]

Ok sorry if this isn't the right place to ask, but what do i need to do to build one of these? Do i just need to buy the bulbs, holders, and ballasts? How does this work?
What wiring needs to be done?
Again, totally new so sorry

[Anonymous No.999379]

>>999105
> Isn't it possible to have both primary and secondary as parallel bifilar?
That gives you four windings; what are you going to do with the other two?

>>999114
> Would bifilar winding be an option for me if my no. of turns is different between the 2 windings?
No.

The main point about bifilar windings is that it ensures that both coils are identical, both in the number of turns and their placement.

This is significant mainly if the windings are carrying equal-but-opposite currents, and you need the resulting flux to cancel out exactly. E.g. wire-wound resistors are bifilar-wound with one end shorted so that their inductance is minimised.

In some other cases, bifilar winding is just used to simply manufacturing (you only need to wind one "wire" rather than two).

[Anonymous No.999382]

>>999270
You'd say that the parameter (e.g. Vcc) was outside of its operating range.

[Anonymous No.999389]

>>999105
>>999114
Bifilar (trifilar, quadrifilar) winding refers to a winding you produce by using 2 (3, 4) wires in parallel. See that toroid pic for example. The basic bifilar winding has 1:1 turns ratio. Trifilar and quadrifilar allow several other ratios by connecting the windings suitably.
There are other winding styles for reducing the leakage inductance for arbitrary turns ratios, but the general idea is to make the primary and secondary winding see the same magnetic flux as much as possible.

Your first step would be to determine whether you need low leakage inductance or low leakage capacitance. Or maybe you have other requirements, like high isolation voltage.

[Anonymous No.999459]

ESP8266 doorbell guy here

I had no idea why transmitting data would fail until I realized that the arduino just doesn't provide enough current to the ESP circuit, so I tried to get a separate power supply from an usb connector to it instead and the thing started spouting jibberish on the serial connection.

After a bit of googling turns out that the problem may be because of the incredibly high baud rate and that I should update the firmware to fix it

so here I am waiting for some usb flashers so I can do this thing

[Anonymous No.999501]

I would like to split the mono signal from my electric bass into 2 signals, one goes through an effects chain, one should remain clean, after that I'd like to combine them into a mono signal again and blend them together.
What would be the smartest way to do this without adding any extra noise and without making the signal significantly weaker? I know that you combine a stereo audio signal (my dry and my wet signal basically) via 10k resistors, but I suppose I'd just use a normal blend poti for this. Is it also possible to have a switch (3dpd switches are available) to turn this feature on or off? I assume I'll need a box with 1 input and 2 outputs and then after all my signals a box with 2 inputs and 1 output (and the blend poti), but how exactly do I wire this up so I don't get excessive noise or other issues? Thanks in advance for any help, I know basic electronics and guitar wiring and I can solder.

[Anonymous No.999506]

>>999358
How I I better? Not cat certified an lower specs from hat I gather??

[Anonymous No.999513]

>>999501
I know this is a slow board I just did some more research, basically can I just use a Y splitter and have 1 cable go through all the effects into a box with a blend poti and the other directly into the blend poti? Will there be issues with impedance (pretty sure most effects have a pretty high input impedance) or would this work? My amp only has 1 input so I'd like to use the blended together signals directly into the amp.

[Anonymous No.999710]

>>999284
You can't call yourself a computer scientist unless you've built something (like a clock with seven-segment-displays) using digital circuits. How do Americucks even pretend to have any kind of standard??

[Anonymous No.999777]

Can someone explain me a thing? I'm trying to learn about pwm but I'm confused. I know pwm can be used to dim a light (but the wave has to be at a high freq). Also something about being about to produce analog output (values between the high and low). This analog value is [ duty cycle x high ] right? For example if I wanted 2.5v I'd use a duty cycle of 50%?

[Anonymous No.999781]

>>999710
Digital logic is required for EE and Comp Sci. EE's do the hardware. What sort of backwards country makes Computer Scientists program VHDL and shit?

[Anonymous No.999783]

>>999513
Quit asking, you can't do what you're doing with an easy circuit, and especially not a passive one. If you still need to DIY it, look up audio mixers.

>>999777
There is no other value than high and low. Literally google PWM.

[Anonymous No.999786]

>>999783
>PWM

I did. This is what I got.

>Pulse Width Modulation, or PWM, is a technique for getting analog results with digital means. Digital control is used to create a square wave, a signal switched between on and off. This on-off pattern can simulate voltages in between full on (5 Volts) and off (0 Volts) by changing the portion of the time the signal spends on versus the time that the signal spends off. The duration of "on time" is called the pulse width. To get varying analog values, you change, or modulate, that pulse width. If you repeat this on-off pattern fast enough with an LED for example, the result is as if the signal is a steady voltage between 0 and 5v controlling the brightness of the LED.

https://www.arduino.cc/en/Tutorial/PWM

>simulate voltages in between full on (5 Volts) and off (0 Volts) by changing the portion of the time the signal spends on versus the time that the signal spends off.

[Anonymous No.999792]

>>999786
>is a steady voltage between 0 and 5v controlling the brightness of the LED.

that part is completely wrong. it's not a steady voltage, it's a pulse. it ACTS as if it were a steady voltage because your eye integrates the pulses of light. that's why it has to be done at a frequency over, say 100Hz, so you dont see any flickering.

[Anonymous No.999794]

>>999792
I get that. I just don't understand how a pulse of 5v can be a value in between 0v and 5v just because of how long it's supplied. That is what the arduino website is saying right? High frequency pluses of 5v can look like a lower voltage? For some reason it blows my mind.

[Anonymous No.999796]

>>999794

there's nothing mind blowing about it. it's just simple mathematical averaging. if you run a light bulb at full power for 50% of the time, it uses 50% power and looks 50% bright.

[Anonymous No.999797]

>>999794

there's nothing mind blowing about it. it's just simple mathematical averaging. if you run a light bulb at full power for 50% of the time, it uses 50% power and looks 50% bright.

same thing would happen if you were driving a car, and instead of pushing the accelerator down 50%, you pumped it down all the way 50% of the time. you'd get the same speed.

[Anonymous No.999798]

>>999797
I dunno why I didn't get it but that analogy helped a lot

[Anonymous No.999800]

>>999786

Key word being "simulated". There are ways of filtering it to make it actually the DC voltage it equates to, but it isn't true DC.

Consider a capacitor (and a resistor, to limit current) hooked up to a PWM supply. The capacitor will not charge to the average DC voltage, it will charge to the peak voltage of the waveform (technically to a limit that approaches the peak).

Or consider attempting to drive an LED with some absurdly high voltage voltage source such as rectified mains. At typical PWM frequencies, it would blow up within the first few cycles, even if your _average_ voltage is too low to even turn the LED on.


But, with some low-pass filtering (say an inductor or RC network) the voltage averages out to be a percentage of the peak value equivalent to the duty cycle. You had it right with your example in >>999777. 5V at 50% is 2.5V, 10V at 70% is 7V, 12V at 30% is 3.6V, etc. That's really all there is to it.

[Anonymous No.999801]

>>999798

Much as I think the water analogy is flawed...consider a pump.

If you leave the pump on full blast, after a minute, you'll have 1 unit of water.

If you drove the pump at half speed, after a minute, you'll have 0.5 units of water.

Now, if you just turned the pump on for half a minute, then off for the other half, you'd ALSO get the 0.5 units of water at the end. It doesn't matter how often you turn it on/off, either. As long as it's on 50% of the time, you'll get 50% of the water.

It works same way with voltage. Well...almost.

[Anonymous No.999813]

>test tl499a boost converter
>sort of works
>only running at 6khz for some reason
>shitty ebay inductor shorts while i'm measuring
tfw

[Anonymous No.999816]

>>999801
>>999800

this more or less answered all my questions. Thanks

[Anonymous No.1000078]

>>999781
>What sort of backwards country makes Computer Scientists program VHDL and shit?
VHDL is too much, but some basic digital circuit theory is usually taught in their computer architectures class.
Maybe he is a CE, he just doesn't know how to translate it to english.

[Anonymous No.1000084]

>>1000000

[Anonymous No.1000108]

> tfw you bought an oscilloscope but you really needed a logic analyzer

[Anonymous No.1000128]

Alright lads.

Studying for amateur radio exams and just electronics in general.

I'm still having problems getting my head around the fundamentals of electricity.

I feel like each video or article explains it differently or not enough.

Any recommendations for youtubers or websites about it?

[Anonymous No.1000206]

What equipment should I buy to get started in electronics?

[Anonymous No.1000214]

>>999710
> You can't call yourself a computer scientist unless you've built something
You're thinking of Computer Engineering. Computer Science (aka Software Engineering) is applied mathematics.

In the UK, CS courses typically have some hardware-related modules, but a) they're usually optional, and b) they're just theory; you don't actually build stuff.

[Anonymous No.1000286]

Alright, I've got a plan to step up my CAD game with some things I have no experience using.

I've got one of these little bastards, and one claim of the thing is that it can be programmed to function like a USB input device. What I'm thinking of doing is making a cube with each face, corner, and edge functioning as separate buttons (For faster camera control in CAD software), and make pressing those buttons mimic a key combo mapped to a specific command.

How would other anons go about making buttons for such a beast? Something non-mechanical would probably work best, but at the same time I want the thing to be completely powered over the USB cable so I don't know if the low voltage is going to cause an issue.

[Anonymous No.1000288]

>>1000286
Dammit, forgot picture

[Anonymous No.1000456]

>>1000206
Read some material from the OP, books will tell you what to get even though they usually go for the "might need at some point" = "get it" mentality that, while not wrong per se, isn't the best way to solve problems especially when it comes to spending money
IMO only things you really need are a breadboard to experiment, a soldering iron and solder to desolder components from old stuff you don't use / get from people that do not use it (a desoldering pump or desoldering wick are not needed and you can always get them later when you got a grip on things), a multimeter (any cheap one should do, I got mine for 7€ and it's great for what it does), solid copper wire for lengthening the component leads to fit in the breadboard and someone to teach you the basics of soldering and desoldering (Internet tutorials will suffice)
Also you -WILL- need to know the theory, from electrostatics to circuits, at least the basic ones (people on YT make great introductions on the matter and you can expand on Wikipedia and other sites)
In particular do a lot of exercises on Kirchoff's and Ohm's laws, you should be able to at least predict voltage levels, current draws and current paths just by imagining a circuit
TL;DR start from the theory then buy a decent soldering iron and get scrap components with that

[Anonymous No.1000509]

So I've got two guitar pedals that I want to run in parallel and then sum up the signal in mono, what's the simplest way to get done with this?

[Anonymous No.1000551]

Is it possible to do via stitching in KiCad?

[Anonymous No.1000611]

>>1000509
You can do it passively (a pair of resistors, or a potentiometer if you want a variable mix ratio) provided that the output impedances of the pedals are much lower than the input impedance of whatever you're driving (which is probably the case).

The resistor values should be significantly smaller than the input impedance (to avoid attenuation), and significantly larger than the output impedances (so that the total resistance of each leg is dictated by the resistance and largely independent of the pedal's output impedance).

If you can't satisfy that constraint, then you need to add a buffer amplifier to provide the desired high input impedance.

Given the schematic in pic related, you have
C=(G1*A+G2*B)/(G1+G2+G3)
where G1=1/R1 etc.

If G3=0 (i.e. infinite input impedance), then C=(G1*A+G2*B)/(G1+G2), i.e. C is a mix of A and B in the ratio of their conductances. If G1=G2=G, then C=(A+B)/2.

[Anonymous No.1000680]

>>999816
>>999800
DO NOT TRUST 800. Capacitors and shit do not work that way at all. If you have a capacitor and a small load, you would still reach peak voltage and act like a normal supply. If you had a small capacitor and large load, you would be reaching the ideal PWM. Using an inductor would even likely blow up your circuit.

>>999801
This anon has it correct.

Put in a mathematical way, changing something's voltage (and current) changes it's instantaneous power use, acting in the "power" domain because you assume the device is on all the time. PWM acts in the time domain, changing how frequently something is active, so the instantaneous power is the same, but the average power is different.

>>1000288
All those digital pins could go between a switch and Vcc, presuming that low is the default state. Then just poll those pins. I mean you have touch pins, idk if those work, but reading digital pins, and you have a lot of them, should be stupidly simple. I didn't say efficient, cause I'd just run something that scanned every 10ms or something if a button was pressed. For buttons, they literally make arcade buttons. There's also an older simple method: two foil sheets with a very thing padding between them. The padding has a hole, so if you press down, the sheets contact. Usually the base sheet is raised slightly, so you don't completely flatten your padding. Literally how DDR pads work. There's also capacitative button chips, which you can of course find for DDR pads.

[Anonymous No.1000684]

>>1000611
The problem with a mixer like that is you're effectively grounding each channel to the other. For AC, every voltage source acts like a ground. If he had two effect pedals, the majority of his signal would feed between them, possibly blowing up both sides. Even if they could somehow handle the sinking, they would heat up and he'd get next to no signal out. Getting tired of this shit.

[Anonymous No.1000685]

>>999813
>buy high power leds for a lamp, 9V
>fragile connections and connected to a metal heatsink
>testing them
>serial 18V, when one shorts it takes out the other
>get more
>didn't notice they were 30V

[Anonymous No.1000702]

>>1000684
> If he had two effect pedals, the majority of his signal would feed between them, possibly blowing up both sides.
Um, no. What do you think those resistors do?

You do realise that an active mixer is simply that circuit connected to a buffer amp (so that you don't need to worry about the input impedance of whatever you're driving), right?

> Even if they could somehow handle the sinking, they would heat up and he'd get next to no signal out.
Oh, you don't.

> Getting tired of this shit.
I'm getting tired of people who don't even have the first clue giving advice here.

[Anonymous No.1000759]

Can you really get an ARM micro with all this shit?

> SPI
> PWM channels
> timers
> UART
> brown out detection
> an ADC -AND- a DAC
> 120 mhz clock
> capacitive sensing?
> "custom logic"? (some kind of FPGA feature?)
> a fucking amp?

[Anonymous No.1000762]

>>1000759
most of your list is standard on uC
amplifier? wont be a power amp, cant plug your guitar in it will be some small signal amp for adc or some shit.
filters and lcd are possible
wireless sounds like a terrible idea
hv fets? fuck off
custom logic? like what firmware? fuck off too.

[Anonymous No.1000800]

>>999325
>Alright niggers: is a fluke 115 a damn good meter or nah? Worth the money? I don't want a piece of shit like I've dealt with up to now. Give me to cold, hard, raw, nasty, ragged truth.
A $20 China generic will almost certainly do what you need. What is wrong with the one you have now?

When you start buying industrial-grade equipment you generally get three things:
--greater accuracy (that you won't need, like thousandths of a volt)
--more features (that you won't use, like USB/SDcard/RS232 data ports, or specialized data interconnects to other high-end equipment from the same company) and
--NIST/other certification papers that cost a lot but are of no value to you.

Go ahead and dump a pile of money for a multimeter if you want to, but for typical hobby electronics--it's not gonna work any better than the cheap ones.

[Anonymous No.1000819]

>>1000286
>Alright, I've got a plan to step up my CAD game with some things I have no experience using.
>I've got one of these little bastards, and one claim of the thing is that it can be programmed to function like a USB input device.
What you have (the picture) is a Teensy 72 MHz Cortex-M4.
It will work, but this is WAY overkill for making a PC input device.

If you had asked about making this before you'd bought anything, I'd say to get a Arduino Nano clone.
On aliexpress, the Teensy that you have costs around $20 and a Nano would cost $1.50.
So you might want to save this for something else more demanding and get a cheaper board for your cube-input project.

-------

I have seen many times that when people who aren't familiar with Arduino-type stuff ask about what board to get to build something, they tend to greatly underestimate what microprocessors are capable of. Most of the stuff that [non-electronics] people ask me about building, only involve a few dozen lines of code and could easily be done on an Uno board. Most of the uses that 'regular people' think up have the processor sitting there most of the time literally just waiting for input.

I usually use Megas for what I build, just because of the greater I/O lines.
The longest program I have written for one so far is about 3500 lines of code, and it still compiles down to only about 20% of the available memory space used. That code might be too big to fit on an Uno, but it would be very difficult for me to write a program so complex that it couldn't fit on a Mega....
And most of the lines in that program is to drive a 7" multi-screen touch-screen interface; what the thing actually *does* is only maybe 5% of the total.
Eventually I will change that setup to PC-slave control, and all the screen display code will get taken out.

[Anonymous No.1000822]

>>1000819
I opted for the Teensy moreso because I knew it had more than enough inputs for what I needed (Originally thought I'd need a shitload of inputs, then realized I could probably eliminate most of them with OR gates)

I've already lost 1 laptop to cheap clones made by fly-by-night companies. Though from what I was looking at, I can use the "Touch Sense Pins" to basically work as capacitive touch sensors as is, so I wouldn't have to have an IC chip or rig up a bundle of sensors within said cube, just have wire run through my 'touch zones'
I'm about 100% sure I have a 70% idea what I'm doing

[Anonymous No.1000845]

What's a good first soldering iron to get? I want to practice soldering over the summer, but I only have access to my dad's soldering iron from Harbor Freight.

Also, what are some cheap projects I can do to get some practice?

[Anonymous No.1000851]

>>1000845
>cheap projects I can do to get some practice
For initial practice, or if penny pinching is your top priority: solder some wires together and to random salvaged PCBs.
What comes to actual projects, choose something which a) interests you and b) doesn't seem to exceed your skills. Guitar pedals might be worth considering, if you happen to play guitar.

[Anonymous No.1000855]

>>1000851
Thanks for the advice! I don't play guitar, but I'll look around for some projects that are interesting to me and that are at my skill level. I have a couple in mind, but I'm gonna need some practice before I tackle them.

[Anonymous No.1000867]

>>1000822
$20 is a bit high I guess, you can find Teensys for like $15....
And if it's the only one you got and that's all you wanna do it's not a horrible crime. (pic related)
Just FYI tho.

>I opted for the Teensy moreso because I knew it had more than enough inputs for what I needed (Originally thought I'd need a shitload of inputs, then realized I could probably eliminate most of them with OR gates)
I dunno what you mean by 'using OR gates'.
One easy way is to use a CD74HC4067 Analog Digital MUX board. This requires 5 pins (it is parallel-control) but allows you to connect to any one of 16 input/outputs at once. There are also SPI and I2C chips/boards that do the same thing.

The other way is shift registers, the 74HC165 is one such chip. I think these are faster (in the SMD versions) but they are digital-only and input-only.
The output chip is the 74HC595.

>I've already lost 1 laptop to cheap clones made by fly-by-night companies.
The only time I've had arduino-type stuff do damage was when I bought some prototyping shields. One had some solder bridges--one of which connected v+ to ground, and it burned the regulator on a Uno. All the other dozens of clones and boards I've bought worked normally.
The explanation here is simple:
All the boards with real-actual-components on them get tested at the manufacturer, and so bad ones get caught.
The proto shields have no components on them really, just a couple LEDs and buttons, and header pins and empty traces. Most of the bridges were between unpopulated traces..... So they don't get fully tested before they're shipped out.

[Anonymous No.1000893]

>>1000867
My original thought was to have a touch zone for each "button" I would need (Which would be every face, corner, and edge so a total of 26 inputs that way). Then I figured I could instead make a touch zone for each needed area, but send them back to an OR gate/IC or whatever they're actually called and send signals back to just 6 inputs.
ie. I touch the front face, and it goes through the OR gate and the pin it's connected to will read high.
Then I touch the edge between the top face and right face, which is split to 2 OR gates, one going to the pin for the top, one for the right, and the program knows to send a different signal because with both of those 2 high, there's only one place it could be
And touching a corner would trigger 3 faces, which again could only represent 1 position on the cube. Figured the OR gates would be a good way to isolate inputs without having to deal with 26 damn pins

Also the cheapo Uno that fucked up my board backfed through the USB connection when I attached an external power supply to the Vin but was still watching the serial monitor

[Anonymous No.1001010]

What is this component called? Its the dc power port. This one is fugged and id like to replace it but i cant figure out the name of this component

[Anonymous No.1001015]

>>1001010
dc power outlet

[Anonymous No.1001039]

I'm making a /diy/ muscle electrostimulator thing, I need to charge the capacitor to a fixed voltage selected with a potentiometer so I can set the intensity level of the discharge, the discharge can either be pulsed, with a period from 3 seconds to 0.2 seconds, provided by a 555, or be manual, triggered by a button.
The options I'm considering are LM317 or OP amp as comparator to stop charge at reference voltage set by the pot. The cap can be discharged up to 5 times a second, so enough current should flow to allow charge it between pulses, this means voltage dividers are excluded.

Any thoughts?

>TL;DR
Want to set an upper limit to the charge voltage of a capacitor while it's discharged several times a second.

[Anonymous No.1001059]

>>1001010
dc power port

[Anonymous No.1001071]

>>1001010
DC jack, DC connector, barrel jack... Or from Digikey's listing: CONN PWR JACK, CONN POWERJACK, DC JACK, etc.

[Anonymous No.1001170]

>>999367
Check google for how to wire fluorescent lamps, yes, you'll need a ballast and starter for each one and a frame to mount them. You can use LED tubes, they're more expensive but don't require a ballast to work.

[Anonymous No.1001180]

Newbie here and in need of some help.

From what I understand, this doesn't need a battery or something, just the positive of the input jack at the 10k pot and the negative going to ground, which is connected to pin 6 or am I wrong?

[Anonymous No.1001181]

>>1001180
Not going to work without power bro.

[Anonymous No.1001182]

>>1001181
yeah, figured, just was confused by the schematic a bit.

[Anonymous No.1001373]

>>999358
Don't both, "fluke" made in china. You are paying for the name with out the benefits.

[Anonymous No.1001380]

>>1000800
I am a Metrologist and I can assure you not all multimeters are made equal. Some cheap multimeters do not match up to everything you need to start off with. Get an Extech or Reed Instrument. Stay away from Ideal (any hardware store brand) or Harbour Freight junk. Spead at least 40 to 60 dollars, you will be glad you did.

[Anonymous No.1001404]

Heads up, Fry's has a sale on Raspberry Pi 3 B's.

I picked mine up in person today. It's $35 in store which is at or below most online prices when you factor in shipping.

[Anonymous No.1001461]

>>1001039

you're doing it wrong, nigger. what you wanna do instead is charge the cap to some fixed high voltage, say 160V, and use various dropping resistors to select the shock level. this is simpler to accomplish, and the results are more repeatable since the circuit is more constant-current-y than constant-voltage. that's important coz skin resistance varies with time, with probe type, with placement, humidity, and from person to person.

[Anonymous No.1001473]

How would I generate the transfer function for this circuit? I built it and the Bode plot is a nearly ideal low-pass filter but I'm having a hard time modeling it

[Anonymous No.1001487]

>>1001473

it's just a simple RC circuit, with R being the OP-AMP's output impedance, typ. around 100 ohms. so it's a first order low pass filter with the usual formula, fc = 1 over 2 pi R C

[Anonymous No.1001489]

>>1001487
That's what I thought, the class for some reason says the output impedance should be 0 ohms but I don't think that that is correct.

Thank you!

[Anonymous No.1001500]

>>1001487
This is the model they've given us to use for an op-amp and even using R-out not as 0 I"m having a hard time seeing how to get the RC as you say. One wire connects Vin- to the Vout node, and I don't really see from there where the capacitor goes - just after the "dependent voltage source"?
Thanks for your help, visualizing this is giving me a lot of trouble

[Anonymous No.1001527]

>>1001473
the capacitor should be connected between the + terminal and ground, not at the output of the op amp. What you've drawn will not work. If your circuit works, your circuit is right but drawing is wrong

[Anonymous No.1001529]

>>1001489
The output impedance is nearly zero when the opamp's feedback is working as expected.
The operation of the circuit of yours depends on the construction details of the opamp itself (actual output resistance, compensation, etc.) and it typically oscillates, unless the capacitor is either small or rather big.

[Anonymous No.1001549]

>>1001489
A theoretical op amp has 0 output impedance. Obviously this is impossible in practice.

[Anonymous No.1001596]

>>1001473
The transfer function is the identity function, H(s)=s.
For an ideal op-amp, both the input resistor and output capacitor are pointless. And it won't be much different for a practical op-amp.

>>1001549
> A theoretical op amp has 0 output impedance. Obviously this is impossible in practice.
Zero is literally imposible, but for a unity-gain buffer, the effective output impedance will be in the micro-ohms range: if the load draws more current, feedback will just increase the internal output voltage to compensate. The closed-loop output impedance is the open-loop output impedance (tens of ohms) divided by the open-loop gain (typically 1e6 or more at low frequencies).

[Anonymous No.1001618]

>>1000759
Cypress PSOCs, they're quite cheap, you can get a prototyping board for 10$

[Anonymous No.1001642]

What is the easiest way to isolate the voltage sensing circuit on my motor controller? I already have fully isolated mosfet drivers and current sensors but I can't work out a simple and compact solution.

Without isolation I then have to ground the high voltage section to the MCU side to read the voltage divider.

[Anonymous No.1001668]

>>1001642
Various options exist:
- Signal transformers.
- Analogue (split-Y) opto-isolator.
- Voltage to PWM and opto-isolating the PWM.
- Likewise, but PCM (i.e. an I2C/SPI ADC).

[Anonymous No.1001695]

>>1001668
Which has the fewest components? I don't have much space, most of the circuits I found on google add 20+ ICs

[Anonymous No.1001750]

How do I get into RF black magic stuff ?

[Anonymous No.1001755]

>>1001695
> Which has the fewest components?
Signal transformers. But they're bandwidth-limited.

Next is an I2C/SPI ADC (or a 2nd uC) + opto-isolators (I2C requires 2x bi-directional opto-isolators, SPI 4x uni-directional; either of those can be had in a single package).

[Anonymous No.1001916]

>>1001596
>for a unity-gain buffer, the effective output impedance will be in the micro-ohms range

you're confused. supposing what you say is true, then a 1Vdc output voltage connected to a 0.1 ohm load resistor would give you 10A of output current.

that's never gonna happen coz in the REAL world, not your micro-ohm fantasy world, op-amps have output impedances in the dozens, or hundreds, of ohms, which will limit the output current to around 100mA.

[Anonymous No.1001921]

>>1001596
>for a unity-gain buffer, the effective output impedance will be in the micro-ohms range

you're confused. supposing what you say is true, then a 1Vdc output voltage connected to a 0.1 ohm load resistor would give you 10A of output current.

that's never gonna happen coz in the REAL world, not your micro-ohm fantasy world, op-amps have output impedances in the dozens, or hundreds, of ohms, which will limit the output current to around 100mA.

[Anonymous No.1001950]

>>1001921
That's what I observed. I ended up with this Bode plot so basically LPF behavior as the other anon suggested. Thanks!

[Anonymous No.1001966]

>>1001921
> you're confused. supposing what you say is true, then a 1Vdc output voltage connected to a 0.1 ohm load resistor would give you 10A of output current.
No, it would give you saturation. At which point, any analysis becomes meaningless. You may as well claim that an open-loop gain of 10^6 (which is fairly typical) "means" that you get a 1MV output for a 1V input.

For the purposes of >>1001473, we're interested in the small-signal impedance (dV/dI), which is actually micro-ohms (i.e. the output voltage changes by nanovolts per milliamp of output current).

The golden rule of designing with op-amps is: Saturation Does Not Occur. It's the EE equivalent of undefined behaviour in a computer program: something you ensure does not happen, because if it does, your "design" is nonsense.

[Anonymous No.1001967]

>>1001921
Hi. Go and learn the difference between open-loop and closed-loop. Kthxbi.

[Anonymous No.1001979]

>>1000893
>My original thought was to have a touch zone for each "button" I would need (Which would be every face, corner, and edge so a total of 26 inputs that way). Then I figured I could instead make a touch zone for each needed area, but send them back to an OR gate/IC or whatever they're actually called and send signals back to just 6 inputs.
>ie. I touch the front face, and it goes through the OR gate and the pin it's connected to will read high.
The problem I see with this is that you might have issues if you want to be able to signal 2 or 3 sides (by pressing them) as opposed to signaling one edge.
If you made all the faces + edges + corners button areas--then that would avoid the issue, but then you'd need to drive 26 different capacitive buttons. (pic related)
That sounds like a lot, but there are little I2C breakout boards for driving 12 capacitive buttons--look for a "MPR121" breakout board.

>Also the cheapo Uno that fucked up my board backfed through the USB connection when I attached an external power supply to the Vin but was still watching the serial monitor
I have seen where if I have an arduino project that uses external power, it may not run properly when hooked up to a PC at all.
I have to unplug the USB cable, and then power down all the external power supplies for a few seconds before turning them back on to test-run the arduino.
The main issue I've seen related to this is with the TFT resistive touch-screens. After a couple sketch uploads, the touch-sensitive part simply stops sending all on its own.
Everything ELSE works, just the touch-screen function stops.....
After several days of frustration, I found that I have to disconnect the USB and do the power-off-on thing to get it to work again.

[Anonymous No.1001982]

>>1001979
Another option is to use a 9DOF board (a gyroscopic-type movement sensor).
This way you would have ONE button, and just rotate it to the orientation you wanted to effect.

Adafruit has a tutorial page about using their item for PC input-
https://learn.adafruit.com/ahrs-for-adafruits-9-dof-10-dof-breakout/software
but you can buy 9DOF boards cheaper elsewhere

[Anonymous No.1002000]

>>1001966
That analysis isn't entirely correct. If the load impedance is too low, you can't actually form a unity-gain buffer because the load forms a voltage divider with the (open-loop) output impedance.

A more accurate formula for the closed-loop output impedance is:

Rout[cl] = (Rout[ol] / A[ol]) * (1+Rout[ol]/Rload)

which is approximately

Rout[cl] = Rout[ol] * A[cl] / A[ol]

(i.e. Rout[cl] / A[cl] = Rout[ol] / A[ol])

If the load impedance is much larger than the output impedance then the closed-loop gain A[cl] is close to 1 and the simplification (closed-loop impedance = open-loop impedance divided by open-loop gain) is accurate.

None of this helps >>1001473 though, because he's trying to understand a circuit involving an ideal op-amp using simulations where the imperfections dominate (either that, or his original schematic is wrong; the schematic itself is something of a troll, but maybe the whole point is to catch out people who just plug shit into a simulator rather than trying to understand it).

[Anonymous No.1002206]

>>1001755
There are also analog optocouplers which have conditioned inputs and outputs so that they don't really need anything else than input and output side power supplies. There might be ones with internal isolated power supplies, too. They aren't horribly expensive. Broadcom/Avago has several models.
Then there are the old fashioned isolation amplifiers, but last time I checked, they were quite expensive.

[Anonymous No.1002209]

I bought some pots to replace broken ones on an amp. But the new ones have this shit post on it, preventing them from sitting flush.

what is this thing, and can I hacksaw it off? I'm guessing it slots into a hole to prevent the pot from rotating? Should I drill a tiny hole for that in the amp housing?

iron is hot, solder is ready waiting for advice on what to do :(

[Anonymous No.1002210]

>>1001461
The capacitor voltage varies between 1 and 9 volts, so I can't use resistors at the input due the lack of appropriate resistor values, but I could use resistors at the output, I didn't thought about that, thanks!

>>1001750
http://hackaday.com/2016/03/23/michael-ossmann-makes-you-an-rf-design-hero/
For the equipment you'll need some candles, ceremonial robes, innocent virgin blood and a goat, a sacrifice altar is optional, but gives your circuits less unwanted coupling.

[Anonymous No.1002211]

>>1002209
That small piece of metal is used to make sure the pot doesn't moves, it is typically inserted in a tiny hole made on the case.
You can bend it with a pair of pliers or just cut it off, it doesn't really matters as long as you tighten the pot to the case properly, if you want to make sure it never comes loose use a lock washer.

[Anonymous No.1002212]

>>1002211
thanks man! waited forever to get these , and had the night put aside to get the amp working ( if possible ) and ran into this stupid problem as soon as I sat down.

[Anonymous No.1002232]

>>1002209
Stops it from spinning the whole component if you try and turn the knob too far. Drill a hole, or just melt one with the soldering iron.

[Anonymous No.1002315]

Hey /ohm/ first time here. Has anyone taken apart a transformer before? I needed to tear some protective tape off my Yamaha reciver main power transformer, and now i need to patch it back. Should i just use normal electrical tape? Is the factory tape any different?

[Anonymous No.1002320]

>>1002315
The factory tape might be more heat resistant, I would recommend using krapton tape, electrical tape can become a sticky mess with heat and time, but I wouldn't worry too much about it.

[Anonymous No.1002324]

>>1002320
alright thanks, heres a pic of the transformer.

[Anonymous No.1002740]

Hello /ohm/, I'm doing a project where I use an attiny and an optocoupler to drive a speedlight, which is supposed to fire when the signal on the cable is shorted
The problem is I'm having trouble finding information on how to wire the optocoupler: I'm using a 4n34 and I've found pic related online (I know it says cn-17 but the 4n35 has the same schematic)
As you can see there's a pullup resistance on the base, but I've found people putting a pulldown, while others leaving it floating.
I have basic knowledge of NPN transistors but I don't get how this is supposed to work...

[Anonymous No.1002747]

>>1002740
>I don't get how this is supposed to work...
It doesn't make sense. While I can fabricate at least 3 half-sensical explanations for that 10k resistor, the most likely explanation is that it was a result of random clueless experimentation or maybe the author just copied someone else's design incorrectly.

Usually you either leave the base open (so that you get a basic phototransistor output) or connect it to the emitter via a large resistor (to reduce the sensitivity of the phototransistor).

[Anonymous No.1002750]

>>1002747
Thank you
I'll probably go with the latter since I'm using it with a 433mhz transmitter and the interference might be a problem

[Anonymous No.1002755]

>>1002740
If the device is designed to be controlled with a switch, you may not be able to drive it directly with an optocoupler. The transistor's C-E current is typically of the same magnitude as the LED current (e.g. 10mA), i.e. a current gain of around unity or lower. So you may need to add an external transistor to form a Darlington or Sziklai pair.

[Anonymous No.1002768]

>>1002755
ok thanks for the heads-up, I'll do some testing

[Anonymous No.1002784]

>>1002768
There's lots of variation in flash trigger inputs.
Ye Olde flashes had 100-400V on their trigger contact and to trigger the flash you had to discharge a capacitor (internal to the flash) charged to that voltage. TRIACs and SCRs were common choices for triggering such flashes electronically.
Newer ones have varying, but much easier requirements. For example, Canon's 430EX II pulls its trigger contact to around 4.5V and the current required to pull it fully down is 1.6mA. In other words, it looks like an old-fashioned TTL input. 1k resistor from the trigger contact to ground was enough to trigger it.

[Anonymous No.1002803]

>>1002784
Luckily it's a new flash, so it should work, but I know where to start looking if it doesn't
Also, how messy does this look?

[Anonymous No.1002901]

I need help identifying parts for my first project.

Prototyping a line of shifters controlled with a simple 0/up 1/down signal. The shifters are in a straight line, packed tightly together

For the prototype I just want to start with applying some code to a micro controller to make these things raise and lower.

Full size would need approx. 550 in a line about 40" across.

Hence the questions what kind of electronic shifter/relay would fit in that space, and what kind of controller would be good to start with 20-30 for a working prototype?

[Anonymous No.1002943]

>>1002901
I have no idea what you mean by a shifter, nor what you're trying to accomplish. There are shift-registers, of course, but I doubt that's it. Best guess is you wanna make an LED light chaser. These were really cool in 1980. So was Madonna.

[Anonymous No.1002958]

>>1002943
Have not heard of an LED Light Chaser but will look it up.

I should not call it a shifter, that's what I'm used to hearing it referred as in my line of work, but it's really just a thin metal plate. About 1/8" thick, 500 in a row and controlled in an archaic fashion. Used for making patterns.

In its most basic form, I want to be able to raise and lower a metal bar with a signal of 0 or 1, and do this with multiple bars at the same time to help modernize and reduce the amount of work in the process.

What I'm not sure of is what kind of switch or motor or actuator would work, and if that is a stock piece or if it would need to be custom made.

Thank you for any help.

[Anonymous No.1002971]

>>1002958

well, solenoids do exactly what you want, but wiring 500 of them would be a nightmare in terms of cost, power requirements and space requirements. they tend to be walnut sized, at least.

so i'd think in terms of having a single solenoid that scans along the line of bars, pushing up those that are meant to be up. or, instead of a solenoid, a motor with a rubber wheel to move the bars in either direction. coz a solenoid only pulls one way, you'd need springs, or some other mechanism to move the other way.

to move the arm, you'd use a stepper motor (like you see in 3D printers, or 2D CNC machines) and either a belt or a worm screw arrangement. the great thing about stepper motors is they do not accumulate positional errors as they move back and forth.

[Anonymous No.1002974]

>>998973
Hey /ohm/ ...

I've gotten a lot of electronics info done quite solidly uptil now, but I still am confused abotu what exactly the energy of a signal is.

I get the concept of course ( the definition is simple enough ), but I can't find a lot about it on the net. Specifically, I'd like to know how to calculate the energy of periodic sine and sawtooth waves.

Also, could someone explain to me why the signal-to-noise ratio formula is 6 * N + 1.76 dB where N is the number of bits your ADC has?

I'm interested in this because I'd like to increase my understanding of noise and data-acquisition.

Thank you in advance!

[Anonymous No.1002977]

>>1002971

Don't think a single solenoid would work, they all need to be raised/lowered simultaneously otherwise it would be much slower than the current process.

[Anonymous No.1003007]

>>1002974
> I get the concept of course ( the definition is simple enough ), but I can't find a lot about it on the net. Specifically, I'd like to know how to calculate the energy of periodic sine and sawtooth waves.
Integrate the square of the amplitude over a cycle.

For a sine wave, sin(t)^2 = (1-cos(2*t))/2, i.e. it's a sine wave which varies between 0 and 1, so its average is 1/2. Thus, the mean power is half the peak power (which in turn is proportional to the square of the amplitude).

For a sawtooth wave, f(x)=x for 0<x<1, integrating f(x)^2 gives x^3/3, so the mean power is 1/3 the peak power.

> Also, could someone explain to me why the signal-to-noise ratio formula is 6 * N + 1.76 dB where N is the number of bits your ADC has?
The 6dB/bit (6*N) corresponds to a four-fold increase in SNR per bit (10*log10(4)=6.0206dB. The mean quantisation error (i.e. the amplitude of the quantisation noise) halves for every bit, power is proportional to amplitude squared so the power of the quantisation noise reduces by a factor of four.

The 1.76dB corresponds to 10*log10(3/2)=1.7609dB. This stems from the fact that quantisation noise looks a lot like a sawtooth wave with mean power 1/3. If the signal is a sine wave (mean power 1/2), then the base SNR is (1/2)/(1/3)=3/2.

IOW, signal power / noise power = (3/2)*(4^N).

[Anonymous No.1003015]

>>1002977
In that case, you'd need 550 solenoids. Which in 40", gives you ~1.85mm pitch. So they'd need to be staggered (you can't get solenoids that narrow; ~10mm/0.5" is probably around the minimum).

One option is latching push-pull solenoids, so that you can use a matrix arrangement to drive them, rather than needing a separate logic output (and transistor) for each one.

A separate output for each one is more expensive, but the solenoids themselves would still dominate the cost (which isn't going to be cheap; each solenoid will be a few bucks, not cents).

In terms of the electronics, it's just a matter of figuring out how to drive one solenoid then replicating it 550 times. You can get as many logic outputs as you like by attaching GPIO chips to an SPI/I2C bus. Each one will need a transistor (BJT or MOSFET; a solenoid draws more current than the logic output will supply) and either a flyback diode or a snubber network (coils generate voltage spikes when you cut the current).

[Anonymous No.1003052]

>>1002901
>Full size would need approx. 550 in a line about 40" across. ...

>>1002958
>I should not call it a shifter, that's what I'm used to hearing it referred as in my line of work, but it's really just a thin metal plate. About 1/8" thick, 500 in a row and controlled in an archaic fashion. Used for making patterns. ....

To get 550 solenoids in a row only ~40 inches wide, the solenoids you would need for this would need to be only about 1/14th of an inch wide,,,, not one-eighth. You don't need to have them side-by-side, you could use levers to spread them out.

Not impossible, but would cost some money. Or some time to make it yourself (with a milling machine...). Also it would depend a lot on how heavy of a mechanical load the rods were.

What is this supposed to be when it's finished?
I would guess a lace/weaving/other needle-craft machine?

[Anonymous No.1003310]

For anti-vibration you usually see rubber washers in-between fasteners. Works great for screws.

What would you do for pic related? You stick a PCB through that spike.
If it's too difficult I'll just give up. It's not that important, I just thought it would be nice.

[Anonymous No.1003325]

>>1003310

Best I can think of is to put an oversized hole on the PCB and stuff a rubber grommet in it.

[Anonymous No.1003345]

>>1003310
Paint it with plasti-dip or wrap it in electric tape.

[Anonymous No.1003428]

>>1003325
>>1003345
rubber grommets sound cool. Didn't know they came in mini sizes, but they're all much too large.

Looks like the solution is going to be the more /diy/ plasti-dip or electrical tape

[Anonymous No.1004034]

I have an inkjet printer. am I still able to print for pcbs? or do i need a laser.

[Anonymous No.1004045]

>>1004034
You can use the photographic method with an inkjet printer. That is, print on a transparency, expose presensitized pcb through it.

Some inks are more etch-resistant than the others and direct printing on copper isn't totally impossible with ordinary ink jet printers, but apparently it isn't easy. Or at least I'd assume it would be a popular method if it worked well and was cheap.

[Anonymous No.1004139]

Hey I made this basic emitter follower based +/-15V rail splitter to use in various projects when I need positive and negative voltages and only have a positive DC supply. I'm a bit surprised it worked to be honest, both in simulation and on a breadboard. I've used op-amp virtual ground drivers in the past and just plain old resistive dividers but the resisitive dividers are unreliable with changes in load and the virtual ground driver I test got really hot and blew up my op-amp so I don't think it was a good design. This seems better, no component is dissipating too much power and the voltage appears to be stable regardless of the load. Only issue is nailing down the right resistor ratio to get exactly +/-15V but I'm not using this for any precision stuff so if I'm off by 10-20mV in either direction it's no big deal.

Anyway I'm sure I've overlooked something so I'd like some feedback so I can know how to make this better.

[Anonymous No.1004237]

>>1004139
The main problem is that the output voltage will fluctuate with load and the division ratio will fluctuate with input voltage.

You'd get less variation with input voltage if the lower leg of the voltage divider was a 4.3k resistor and a diode (or another emitter-follower). As it stands, you have to tweak the resistance ratios so that the reference voltage is off by Vbe ~= 0.7V. But clearly any resistive divider is going to produce a reference voltage proportional to the input voltage.

You'd get less variation with load current if you reduce the output impedance and/or increase the gain. This is why an op-amp is usually preferable. But op-amps can only supply ~10mA; if that isn't enough, you can use an emitter follower to boost the output current from an op-amp (include the transistor in the feedback loop to preserve the linearity).

Whatever approach you take, you can't avoid the fact that you'll be dissipating half the supply voltage times the ground current (i.e. you're effectively building a linear regulator). Linear rail splitters are designed for the case where "ground" is a reference voltage, not a power rail.

[Anonymous No.1004245]

>>1004034
If you're planning to use the 'toner transfer' method, you need a laser.

[Anonymous No.1004342]

>>1004139
You can buffer the output of the OA driving the virtual ground with 2 emitter followers (NPN for the positive and a PNP for the negative output).

[Anonymous No.1004429]

Being relatively new to electronics, is it worth it to salvage components from broken electronics?

[Anonymous No.1004500]

>>1004429
> Being relatively new to electronics, is it worth it to salvage components from broken electronics?
Rarely. And if you're new to electronics, you probably won't know what's worth salvaging and what isn't.

There's no point in salvaging stuff which
- you're unlikely to find a need for
- which can't be identified
- costs a few cents to buy new
- is likely to have failed
- you can't use (e.g. SMD parts if you're going to be using stripboard).

[Anonymous No.1004507]

Is there any chance to get any of the SoC manufacturers to give a sample or two?
What's the usual minimum lot size?

[Anonymous No.1004863]

I just noticed there's a maker faire 2 months from now in my city. I kind of want to make something to show off. Figure it'd look good on a resume too.

What's something that I could make in 2 months that'd be fairly impressive? I'm a recent EE grad with about 5 years hobbyist experience focusing in digital logic.

I was considering maybe a tic tac toe game AI made entirely of relays or an etch a sketch that takes people's pictures then sketches them... Idk.. Any other ideas?

[Anonymous No.1004889]

>>1004429
Sure. If the shits broken you might as well take what you can from it. Just do some googling before you go reading into everything so you know what to look for. Theres some parts worth salvaging and some parts not worth salvaging. But in any case taking things apart and looking the boards over for parts to take will get you some practice identifying components and shit. Give you a little insight into how stuff works. So yeah if you have the broken electronics, why not crack em open and see if you can salvage some shit?

[Anonymous No.1004893]

>>1004507
Well, TI offers some of their SoCs as samples, but those tend to be available from distributors like Digikey as well and TI has gotten increasingly faggy with their sample policy lately - which is why you shouldn't request shit just because you can.
Typical minimum order (when you can't buy just one) is one waffle or reel. Some of the aforementioned TI SoCs were available as waffles of 126.

[Anonymous No.1004909]

>>1004863
> there's a maker faire 2 months

You're not going to get a table/booth to put you stuff on if the registration deadline has passed, and it probably has. You can still show off by building something wearable and wandering around as a regular attendee. A helmet decorated with LEDs, a nixie-tube watch, shoes that play DTMF tones as you walk, etc.

[Anonymous No.1004918]

>>1004507
Sure but you do realize you need to make a pretty knarly PCB?

[Anonymous No.1004938]

>>1000800
Fluke meters have the one thing over Chinese meters you forgot to mention, longevity. They are also less likely to wander off specs as quickly as a Chinese type will so should be more accurats for a longer period b4 needing recalibrated.

[Anonymous No.1004981]

>>1004909
It hasn't passed. It's a community 'mini make faire' so nothing official. The deadline is soon though so if I want to do something I need to choose soon.

[Anonymous No.1005021]

>>1004893
That's cool. I can afford to buy a waffle if it means the order size would be in low hundreds ~10$ each
I honestly thought that I had to pay for 1000 or more.. which is way above my pay grade

>>1004918
That wouldn't be a problem. I thought I just design the thing and have a hundred or two printed for me. I'm a CE student and know a little electronics as well.
I have BGA work station and can solder the components myself.

Thinking of starting a small business if I can make a Return of investment. that's a year or two down the line ofcourse

[Anonymous No.1005024]

>>1004981
> It's a community 'mini make faire' so nothing official.

The term "Maker Faire" is owned by Make magazine. An event can't be called that without paying them a licensing fee, getting their approval of the venue and security, following their rules on what projects are allowed, and a lot of other official stuff. "Mini" doesn't mean the event is going to be small. They double the licensing fees unless events are called that.

[Anonymous No.1005036]

>>1005024
It is done by make, but it's not as big as the Kansas City one the month prior . This is the faire I'm considering:
https://makerfairewichita.com/

The deadline for registration is tomorrow.

[Anonymous No.1005072]

>>999180
Im laughing at this. My compsci fundamentals class was 100x harder than making a clock. We had to use C on bb gpio connectors as this ass wipe makes a clock. What a time to be alive.

[Anonymous No.1005083]

>>1005072
>see this post
>turn around and look at my fpga loaded with the risc cpu architecture I hacked away in 2 weeks for lousy 2 grade points on a 20 scale

And here I thought we (software engineering) had it easy

[Anonymous No.1005237]

>>999180

>be mechanical engineer
>like electronics
>decide to make a Nixie clock for shits and giggles

Meh, it's not that rough. Still doing the Eagle sketches, but I had a working prototype and know what I have to do.

[Anonymous No.1005271]

I'm a mechanical engineer, and I don't do too much with electronics, but I do know the basics.

In my dumpster diving escapades I have run across a bag with about 100 mosfet drivers and like 6 mosfets.

I know how mosfets work, but what the hell does a mosfet driver do? I looked them up but all the explanations are over my head. Are these things useful or just a meme?

[Anonymous No.1005287]

So. Let me get this straight.

AC - no net flow of electrons over the circuit, but taking current to be change in charge in coulombs per second, then electrical current is constantly flowing through a given cross section of a circuit. Right?

Additionally, one shouldn't think of current in a DC circuit as moving electrons, but a raised quantum state of electrons in a given part of the circuit Passing energy to the next. There is no net flow of electrons, they are fucking doing whatever they want in the crystal structure of the metal.

[Anonymous No.1005294]

>>1005083
Developers Developers Developers Developers
Developers Developers Developers Developers

[Anonymous No.1005321]

>>1005271
> I know how mosfets work, but what the hell does a mosfet driver do?
There are two sorts: low-side (where the source is connected to ground and the drain is connected to the load) and high-side (where the drain is connected to the positive rail and the source is connected to the load).

A low-side driver is typically just a current amplifier. High-power MOSFETs have significant gate capacitance. In order to switch them quickly (and thus minimise the time spent in the linear region where they're dissipating significant power), you need to charge and discharge that capacitance quickly, which means high current (typically several amps).

A high-side driver has the additional problem that when the MOSFET is saturated (turned fully on), the source voltage is (ideally) going to be very close to the drain voltage which is the positive rail (this may be well above logic levels), and the gate needs to be a few volts above that. So as well as providing high current, they need to level-shift (i.e. supply an output voltage which is much higher than the input voltage). And they often include a voltage doubler to generate a voltage above the positive supply rail.

And because one of the most common uses of MOSFETs is for a half-bridge (push-pull) with both a low-side and a high-side MOSFET, it's common to have driver ICs which include both a low-side driver and high-side driver.

[Anonymous No.1005325]

>>1005287
> Additionally, one shouldn't think of current in a DC circuit as moving electrons
In solid conductors (and thermionic valves), a DC current corresponds to a net flow of electrons. The net flow is very slow compared to e.g. the orbital motion, but it exists.

In ionised gases (e.g. flash tubes) and ionic liquids (e.g. electrolysis), current is also carried by the flow of positive ions (and sometimes negative ions) as well as by the flow of electrons.

[Anonymous No.1005333]

>>1005287
>electrical current is constantly flowing
Well no, you have nodes. At the same time, the electrons are moving for 1/2 wave in the same direction, so at 1amp rms, 60hz and 80% speed of light, the electrons have moved 2,000 kilometers.
>in DC there is no net flow of electrons
You are literally wrong. Movement and amps are not measures of accumulation, they are measures of flow.

[Anonymous No.1005334]

>>1005333
I'm mistaken on the first part. The power is transmitting that fast, the electrons are moving much slower. The second part is 200% true.

[Anonymous No.1005352]

Why is it that a resistor is normally put on the anode of an LED? I looked it up to see if there is a difference, and apparently there isn't. Is it just because the cathode is shorter, and the resistor is easier to attach to the positive side?

[Anonymous No.1005357]

>>1005352
If anything, it might be a retarded human need to have the "last thing" in the circuit be the one that is doing "stuff", i.e., the "useful" parts of the circuit should be the closest to ground. There is no reason, there is no convention, don't try and read into it.

[Anonymous No.1005358]

What should I put on this cubesat?

[Anonymous No.1005438]

>>1005333
Amps are a measure of movement of charge. Charge doesn't necessarily mean electrons moving senpai.

[Anonymous No.1005461]

I have been looking at these AMC1200 isolated op amps for current sensing but I don't understand how to use them. My MCU software expects a single analogue input that sits at 1.665 when there's 0 current, 3.3 at max current and 0v at max reverse current.

But all the datasheets have this ICs dual outputs going to the ADC, I don't want to change the software, how do I turn it back into a single out put that varies based on current direction?

[Anonymous No.1005466]

>>1005461
>how do I turn it back into a single out put that varies based on current direction?
By measuring only one of the amplifier outputs. The datasheet says you'll get (around) 1.29V at zero current and 1.29V +/- 1V at +/- full current.
So either you change your SW or add extra amplifiers to change the scaling and offset. If you used two ADC inputs (or a real differential input), you'd get rid of large part of the offset drift.

[Anonymous No.1005476]

>>1005466
Could I just place op amps like this so I get a single signal with the full 0-3.3v range? I don't think I have enough MCU pins to consider 7+ differential ADC inputs.

[Anonymous No.1005484]

Guys, would running 25.2V through the 12V Relay LOAD side NOT the coil side, blow the relay if the current is under the rating?

Picture very related, would it blow the 12V rated relay

[Anonymous No.1005486]

>>1005484
what's the current? that voltage probably not.

[Anonymous No.1005487]

>>1005484
Just to clarify I want to use 25.2V through the 12V relay switch side/load NOT the coil, the coil would be powered by 12V

[Anonymous No.1005489]

>>1005486
10A max current.

[Anonymous No.1005492]

>>998973
i want to set up an electronics repair workstation on the cheap, any ideas?
i need to know what brands and models to look out for, which brands and models to avoid, where i can cheap out, and where i cant, as well as what equipment ill need.

[Anonymous No.1005496]

>>1005476
That would be a ridiculous overkill and AD8130 does not have rail to rail output (and needs 5V supply), but yes, something similar would do the job.
If this was my problem, I'd seriously consider if could relax the requirements for the ADC input voltage range and if that's not possible, if I could use a more convenient current sense amplifier.

>>1005484
If the contacts are rated for 12V only, then you can expect that the relay dies faster than normally. In the worst case, you'll get a permanent arc between the contacts and the relay melts. No idea how likely that is in reality.

[Anonymous No.1005503]

>>1005496
I will then have to place the relay after the resistors in my closed loop which would make the voltage at safe level

[Anonymous No.1005505]

>>1005484
> 12V Relay
Coil voltage and contact voltage are different things. If you want to switch 25V, check that the contact voltage is >25V DC. Relays designed for switching mains typically have contacts rated for 50V DC.

Note that your circuit doesn't show a load: the relay contacts are shorting the battery.

[Anonymous No.1005510]

>>1005476
You could but you can just use the non-inverted output (108dB of common-mode rejection should be sufficient), and you don't need an instrumentation amplifier; a plain op-amp used as a differential amplifier would suffice.

But using an extra analogue stage just to avoid changing the scale and offset in the software seems a bit silly.

[Anonymous No.1005514]

>>1005505
The real circuit has a load. The circuit is a lithium polymer 6S discharger for when my 1/8 RC car crashes or if the weather decides to rain and i have to drain it back to storage level quickly. The relay is just to add a on/off switch. I am very stupid with this stuff and i am sure it can be done better but whatever lol

[Anonymous No.1005567]

100 or so 555 timers. Wat do?

[Anonymous No.1005612]

>>1005567
Hackaday did a 555 contest a few years ago, look it up and have fun

[Anonymous No.1005739]

Got a cheaply made USB power bank thing. Any reason I can't swap the 18650 inside it for something higher capacity? Current one in it is totally unbranded and only charged my 2600mAh phone battery 20% higher.

>>1005612
Thanks.
https://web.archive.org/web/20120918091914/http://www.555contest.com/

[Anonymous No.1005988]

/ohm/
I left my motorcycle keys inside a trunk on my bike which is opened only by the keys.
The keys activate a switch which powers a 12v solenoid
Without the keys I would need an external 12v capable of going through about an inch of foam and plastic

I don't think a Van de Graaf is gonna cut it, what do?

[Anonymous No.1006013]

>>1005988
Are you joking or something?
Remove foam and plastic Or call a locksmith.

[Anonymous No.1006136]

I've got a bunch of Xilinx XC9536 chips and the programmer for them. They're CPLDs. Useful or no?

[Anonymous No.1006140]

>>1006136
If you need rather small 5V CPLDs then yes. If not, then no.

That said, chips which have about the same number of macrocells and IO pins tend to be too small for my uses.

[Anonymous No.1006161]

Hi /ohm/ it's summer again and my laptop is running hot as a toaster, reaching 70 to 80ºC without load.

I checked the fan's PWM and it's always running at less than 50% of it's capabilities despite these temps. I bridged the PWM and power pins with some tweezers and it immediately reaches full speed.

I was wondering....
a) Can I change the PWM with software or settings to make it run colder?
If not:
b) Can I make my own nigger rigged PWM control without killing the board? My plan is to add a multivibrator or a 555 CMOS and a pot or "turbo button" at one side to go from 50% to 100% duty cycle.
I'm guessing I'll have to cut the PWM wire to avoid damaging the original driver with the new signal.

Video of the test:

https://www.youtube.com/watch?v=qy5NOvIjppQ

[Anonymous No.1006167]

>>1006161
Replace the thermal paste and use a software solution. Check the BIOS first, then try speedfan.

[Anonymous No.1006178]

>>1006167
Thanks I already had checked the BIOS but there was nothing in there I could use, I'll try speedfan, although replacing the thermal paste might be a better idea.

[Anonymous No.1006311]

This is probably a really stupid question for this thread, and I'm sorry. I'm fixing up my old receiver, and I'm replacing the bulbs in the meters with LEDs, and I can't get all of them to light up (the dial pointer one has a dim glow to it). It appears that all of the lights are wired in series, so I put a single resistor on the positive side of the dial pointer LED.

What is the problem with how I am wiring them up?

http://akdatabase.org/AKview/albums/userpics/10004/Yamaha%20CR-420%20Schematic.pdf is the schematic of it

[Anonymous No.1006323]

>>1006167
Well, speedfan did not recognize my fan nor was I able to change the PWM, I'll try to change the thermal paste before trying to make a controller.

>>1005492
https://www.youtube.com/watch?v=bVoBZqlvCKo

>>1005739
>Any reason I can't swap the 18650 inside it for something higher capacity? Current one in it is totally unbranded and only charged my 2600mAh phone battery 20% higher.
Sure, give it a try, most chargers have generic boards to charge and discharge the bank battery with proper voltage limiting and so on, their operation doesn't depends on the battery's capacity.

>>1006311
Are you wiring the LEDs in parallel, if so you must use a resistor at each LED to avoid current unbalance, if you're wiring them in series make sure to use the same kind of LEDs, the intensity and color can vary depending on the manufacturer even if they're all sold as "cool white", for example.
Use this tool:
http://led.linear1.org/led.wiz

[Anonymous No.1006330]

>>1006323
senpai, i dont have several grand to drop on this

[Anonymous No.1006542]

>>1006311
>I'm fixing up my old receiver, and I'm replacing the bulbs in the meters with LEDs, and I can't get all of them to light up (the dial pointer one has a dim glow to it). ...
>What is the problem with how I am wiring them up?
If I was doing this, what I would do is use a buck converter to run the 12 vdc down to about 4 ~ 5 volts, and then wire all three LEDs off that in parallel, with each having its own dropping resistor.

[Anonymous No.1006590]

So I'm building an Arduino controller with encoders, but the cheap chinese ones were really shit, so I bought 400ppr ones. It says 5-24V in specs, but I've read that you need about 9-12V for them to operate normally. What is the best way to power both encoders and arduino from a single USB connection?

[Anonymous No.1006690]

>>1006590
Power the Arduino directly from the USB's 5V, (although it isn't guaranteed to be above 4.4V for a bus-powered hub) and use a boost converter to generate e.g. 12V.

A boost converter can be made with a dedicated switching regulator IC (e.g. MC34063) or using the Arduino. Or you could buy a self-contained DC-DC converter module.

The main thing to bear in mind is that USB imposes limits on both the stable load current (100mA until you successfully negotiate a higher limit) and the load capacitance (10uF). And you're only allowed 500uA when suspended (so don't try to run the boost converter during suspend).

[Anonymous No.1006691]

>>1006690
So if I buy separate 12V adapter, can I control it with logic-level MOSFET?

[Anonymous No.1006745]

What's a good project that I can make out of shit pulled out of the trash? I dig through a lot of trash and have some basic programming & electronics knowledge from doing ham radio stuff.

[Anonymous No.1006755]

>>1006691
Yes.

You have two basic options depending upon whether the load goes on the high side or low side (high side is simpler).

[Anonymous No.1006762]

So I just got a vintage pachinko machine given to me.
Its function is fully mechanical, but there are lights you have to power.

Some people are using 24v wall warts, some people are using 9v batteries, and some people are running straight AC transformers.

Im not too knowledgeable, but how can the lights not give a shit about 9v-24v AC or DC?

Is there one that is going to be safer than the other?

[Anonymous No.1006783]

>>1006762
ah I figured out that the 24v machines are special, and most will just run on 9-12v.
Thought it was weird.

[Anonymous No.1006808]

>>1006311
>>1006323
>>1006542
I got the lights in the receiver to work. I had to change the way that one of them was connected.

Here's a picture. Sorry it's blurry. I'm not sure if I'll keep the green light in the dial lamp, since I don't like how it removes the red line. It was white originally.

[Anonymous No.1006827]

What are some good beginning projects that are useful/can be used for more advanced projects? I want to make something but I'm really not interested in the novelty of things without utility.

[Anonymous No.1006835]

>>1006827
12v power supply

[Anonymous No.1006840]

Can someone explain to me how exactly the reset button works?

I'm a beginner at this

From what I gathered, the IC1 (74LS373) is an asynchronous D-latch which gathers input from the pushbuttons.
When a button is pressed, one output goes to LOW and the NAND gates disable the latch (LE, pin 11), meaning no further input. This also goes over the encoder, inverter and decoder to the 7seg display, and a buzzer sounds, but this doesn't matter now.

What I don't understand is how is a second input recorded? Does the reset have to be pressed every time after a button is pressed to be able to receive another button press?
Reset is connected to output enable, /OE, active LOW

The outputs of the IC1 (74LS373) are in high impedance (Z) state when the button is pressed (/OE in HIGH).

How do the outputs on the latch then go back to zero when the button is released (does that happen at all)?

Does pressing the reset button bring the VCC pin of ALL the integrated circuits to GND, therefore ......resetting them and bringing everything in start conditions? i think this makes the most sense but would like someone to confirm

Does the reset have to be pressed after every button input?

[Anonymous No.1006860]

>>1006840
Brings them all to ground, thus resetting the entire system.

[Anonymous No.1006960]

>>1006840
> Does pressing the reset button bring the VCC pin of ALL the integrated circuits to GND,
No. And even if it did ...

> therefore ......resetting them and bringing everything in start conditions?
Cycling the power doesn't reset the IC's state. Typically, the state is indeterminate after a reset.

Most complex logic circuits have an explicit power-on reset signal generated by a R-C low-pass filter on Vcc. When power is applied, a capacitor is charged through a resistor. The capacitor voltage is the reset signal (active low). This is connected to any device which needs to be reset upon power-on.

> Does the reset have to be pressed after every button input?
Yes.

Reset puts the outputs into high-Z which is equivalent to being active high, as logic inputs will float high if disconnected. Although relying upon this is a hack. IMHO, it would be better to connect the reset switch to pin 12 of IC4 (with the logic inverted, i.e. 0 when pressed) and just tie OE to ground.

This causes the output of IC3 (74LS30) to be low (all inputs are high), which causes the output of N2 (IC4, 74LS00) to be high, meaning that LE is high. In this state, the outputs will track the inputs. Assuming that none of S1-S8 are pressed, all inputs will be 1. When reset is released, all outputs will be 1.

As soon as any of S1-S8 is pressed, the input will go low, the output will go low, the output of IC3 will go high, the output of N2 will go low, LE will go low, latching the data.

[Anonymous No.1006972]

I've looked through the booklist and channels, but there isn't a book that's just one breadboard circuit after another which is what I'm looking for, not the complete theory of each component which I already know.
any

[Anonymous No.1007016]

simplest way to get this to become variable speed?

[Anonymous No.1007027]

Quick question:
Can you run LED bulbs (the ones that go into standard lightbulb sockets) on DC? Since most of them use a bridge rectifier I would assume you could just put 120v dc to the terminals and it'd work fine.

[Anonymous No.1007031]

So uh /ohm/, this isn't electronics related but I'm going to ask it anyway... There's a coworker of mine who is also an engineer like myself. I really like her, a stronger interest than I've had in any other girl. How do you approach EEs or DIYers of the opposite gender? Should I invite her to the maker faire or something? I dunno what to do. I figure other EEs might know.

I don't understand women. I spent all of highschool and college in my books and with my iron. Never got any girl experience......

[Anonymous No.1007038]

>>999459
Can't you flash it with your arduino? Using my uno to program my other arduinos

[Anonymous No.1007042]

>>1006960
I didn't know about high-Z being equivalent to high due to pullups on the input in this case, thanks

>> Does pressing the reset button bring the VCC pin of ALL the integrated circuits to GND,
>No.
Is this due to the R9 resistor on the bottom, making not all the current go through reset to the GND and preventing short circuiting the entire thing? All the other IC's continue to work as normal when reset is pressed?

[Anonymous No.1007063]

>>1006972
>breadboard
Sounds like you basically want the companion book of some 100-in-1 electronics kit.
If you can drop the breadboard requirement, then the old hobby electronics books from the sixties and seventies would probably be what you want.

[Anonymous No.1007065]

>>1007042
It worked with the old bipolar 7400 logic families (7400, 74S00, 74LS00, ...) because their "low" was in reality "pull enough current out of the input" and "high" meant no current.
It does not work with the modern CMOS 7400 families like 74HC00.

[Anonymous No.1007078]

>>1007063
that sounds about right. just want one simple circuit after another.

australian, so anyone know if there's one at jaycar?

[Anonymous No.1007116]

>>1007016
It depends upon the motor. Most AC motors are synchronous, i.e. their speed is dictated by the AC frequency, so if you want to change the speed you need a variable-frequency drive (VFD).

[Anonymous No.1007123]

>>1007027
Maybe. But a rectifier and capacitor connected to 120V AC will produce 170V DC (120V AC is 120V RMS, 170V peak). There are other reasons why you may not be able to use DC for devices designed for AC, but most of them are unlikely to apply to a bulb.

But you should avoid dicking around with 120V DC unless you know what you're doing. In particular, switches and relays rated for 240V AC typically aren't rated for more than 50V DC. And the shock hazard is greater with DC.

[Anonymous No.1007131]

>>1007042
> I didn't know about high-Z being equivalent to high due to pullups on the input
It's not pull-ups; a TTL input stage is a common-base amplifier (i.e. the input is connected to the emitter). Pulling the emitter low increases the collector current, reducing the collector voltage.

Note that this wouldn't work if IC3 was replaced with a CMOS (4000-series) equivalent (CMOS inputs don't float high) and would be unreliable with 74HC (these tend to float high but with very high impedance, so noise on Vcc gets coupled to the input).

> Is this due to the R9 resistor on the bottom, making not all the current go through reset to the GND and preventing short circuiting the entire thing?
Pretty much. If the resistor wasn't there, S9 would be a short circuit which could damage the power supply.

If for some reason you wanted to force the Vcc rail low, you'd put a normally-closed switch between the external 5V supply and the Vcc rail, so that pressing it cuts the current, not by shorting the rail to ground.

> All the other IC's continue to work as normal when reset is pressed?
Reset just sends OE high, forcing all 8 outputs to high-Z (effectively to 1 for TTL) The capacitor across S9 acts to de-bounce the switch and also acts as a power-on reset.

[Anonymous No.1007157]

I have an 8 ohm 5W loudspeaker and want to make an amplifier for my phone

I'm using CE and CC stages but idk how to treat a loudspeaker. Is it a resistor in small signal or what?
I need to be able to find the necessary gain but idk how to if the loudspeaker is there

[Anonymous No.1007171]

>>1007131
>>1007065
Alright thanks, i think i finally understand it now

[Anonymous No.1007173]

>>1007157
Usually it is treated as a resistor and the amplifiers are dimensioned so that they would produce the expected output power if the load was resistive. If speaker inductance is taken into account, it is common to just design the amplifier so that the inductive effects won't cause problems (like kill the amplifier or cause excessive distortion). Sometimes inductance is compensated, but to do this accurately requires you to know the used speaker.

The reality is more complex due to the mechanical effects.

>8 ohm 5W loudspeaker
So... To get 5W rms sine output, you'd need 6.3V rms output voltage, or +/-8.9V peak output. +/-12V dual power supply would be attractive.
Except that if you drive the amplifier into clipping, then you're actually feeding 10W (8.9V square wave) into that speaker, probably even more if you're using that 12V power supply. Meaning that you might want to design your amplifier for lower than 5W sine output.

[Anonymous No.1007181]

>>1007157
It's an impedance comprised of resistive and inductive components. If you know the inductance, you can use a Boucherot cell (R+C in parallel with the speaker) to make the impedance real (purely resistive).

But treating it as an 8-ohm resistor is mostly adequate.

To actually get 5W from an 8-ohm speaker requires ~6.3V RMS. At low voltages, you'd normally use an H-bridge (i.e. connect the speaker between two amplifiers with one inverting).

A class D amplifier (i.e. driving the amplifier directly from a sigma-delta signal) has higher efficiency than a linear amplifier.

[Anonymous No.1007197]

How cheap should you go with soldering tools as a complete newfag when it comes to this? Anything specific that is worth/isn't worth buying? I wouldn't be using this every day but I have a few things I'd like to fix up and I thought I might as well learn to do it myself.

[Anonymous No.1007209]

>>1007197
Depends what you're doing really.

Anything surface mount, you can't afford to go cheap.

But if it's just oldschool through-hole stuff then a cheap iron should be just fine, as long as it has a nice selection of tips available.

[Anonymous No.1007242]

>>1007181
how would i go about finding the inductance of the the speaker?

[Anonymous No.1007247]

how do i invert the polarity of a dc voltage?
i have a pc psu im making into a variable 24v power supply for shigs and it can do 25 amps with the +12v lines but only .8 amps with the -12v line. so to make it able to output 24v with a good amount of juice i want to invert half of the +12v lines and use them in place for the projects that use the + and - 12v lines to make the voltage variable.
is this possible and how do i do it?

[Anonymous No.1007297]

>>1007247
>is it possible
No.

[Anonymous No.1007303]

>>1007247
That sounds like a good way to blow up a PSU.
Generally, putting power rails in series (unless the device SPECIFICALLY says they are intended for this behaviour) is going to blow something up.

If you want to output 24V, fucking around with +/-12V is just going to make it complicated anyway, unless you don't give a single shit about proper grounding either.

[Anonymous No.1007323]

>>1007247
> how do i invert the polarity of a dc voltage?
With a switching regulator. Which largely defeats the point. You may as well just use the PSU to drive a 12->24V boost converter.

[Anonymous No.1007661]

>>999781
>>1000078
>>1000214
not that guy but there are (as far as I know) no CE degrees in germany. CS (Informatik) students need to complete at least one module on computer architecture (usually with hardware lab/projects). I'd say CS and EE are roughly interchangeable in the first three semesters here

[Anonymous No.1007683]

>>1007661
> I'd say CS and EE are roughly interchangeable in the first three semesters here
Either you're wrong or the German system is totally fucked up.

EE contains a ton of analogue stuff which is entirely relevant to EE but not even remotely relevant to CS.

[Anonymous No.1007689]

can i get some recs for resistors? Im building pic related, a contact mic preamp. Ive got some old wima mkp10 caps already, but dont really know how to choose resistors. Any guesses on what wattage rating would be sufficient?

[Anonymous No.1007695]

>>1007689
The cheapest, most common through hole resistors are totally sufficient.

[Anonymous No.1008177]

Is this book is worth reading? I know you recommend it for beginners but I am only a few pages in and I feel like a baby. I hate the shallow explanations and I am not even sure if it's worth taking notes.

[Anonymous No.1008178]

Is this book is worth reading? I know you recommend it for beginners but I am only a few pages in and I feel like a baby. I hate the shallow explanations and I am not even sure if it's worth taking notes.

[Anonymous No.1008315]

>>1008178

just google shit you are confused about or if you are still stuck, post questions

[Anonymous No.1008320]

>>1007689
the rainbow peanut kinds

[Anonymous No.1008400]

>>1008178
Haven't seen it, but I'd assume it is comparable to the other recommended beginner books. That is, very simple explanations, plenty of analogies, minimal amount of math.
I don't really know what you're expecting, but Art of Electronics might be what you want (covers quite a lot, has practical focus, uses math when it is needed, but is not obsessed with theory and analysis of castrated circuits).
Or maybe you just want your typical university beginner book, which revolves around the theory and analysis or circuits and explains things with math, but tries to avoid real-life complications.

Or you could read that beginner book as light amusement and then try something else. At least it would make the basic concepts familiar.

[Anonymous No.1008567]

>>1001473

V_L is the same as the output of the amp, so ignore the capacitor. R_In is in series with the infinite input impedence of the opamp, so ignore that too.

You now have a buffer amplifier.

If this is supposed to be non-ideal components then tell us which.

[Anonymous No.1008568]

>>1008315
I am not stuck I just feel like it's a waste of time since the definitions are so basic. I guess I am worried that it will give me a shallow understanding of a topic when I could get an in depth explanation from another book. Anyway it doesn't really matter it's only like 128 pages so I'll read it anyway.
>>1008400
>Or you could read that beginner book as light amusement and then try something else. At least it would make the basic concepts familiar.

Yeah that's probably what I am going to do, Thanks.

[Anonymous No.1008660]

My niggas

Im an EE major but beginner in electronics hobbying. I have a DC power supply and a shit ton of components. Whats some cool shit I can build to start learning this stuff?

[Anonymous No.1008670]

>>999325
Unless you're doing something that requires lab or serious industrial precision, just go with Triplett. Cheaper price, and almost as much longevity.

[Swifts !WiNDUPyWhk No.1008671]

>>1008660
Well.. a power supply is a good project, but you already have one..

Here's a list of projects, rolling in the thread is naturally discouraged.

[Anonymous No.1008673]

>>1008671
I can still build a power supply. Cant hurt and Im sure its a phenomenal project to learn how this stuff all works together.


I wish EE undergrad programs taught more actual circuit design instead of cramming all the theory down our throats.

I can analyze the SHIT out of a circuit, but don't ask me to design something

[Swifts !WiNDUPyWhk No.1008674]

>>1008673
I hope it gets better.
I'm a technician right now, one-upping to Engineer in the following 3 years.
We never actually designed anything or analyzed anything. Just got the basic math and theory down and built a couple of projects.

Atleast we did a lot of work with PLCs. That's something.

[Anonymous No.1008683]

>>1008674
You'll pass engineering school easy. You dont really learn anything but theory with no applications

[Anonymous No.1008688]

>>1008673
> I wish EE undergrad programs taught more actual circuit design instead of cramming all the theory down our throats.
You can't design circuits without the theory.

And design is something which largely needs to be learnt rather than taught.

One reason not to choose a PSU as your first practical project: safety. PSU design is less about how it works than about how it fails.

[Anonymous No.1008690]

>>1008688
Yea I completely agree that you need to know the theory to know how to design. Also, sure, learning to design circuits comes with experience, however a senior level class about intro to circuit design where you learn a few basic circuits and maybe do some sort of project would be great.

I graduate this December and literally only know how to make HP, LP, BP, and BS filters and a couple of amplifiers. Also I know very basically how zener diodes can be used in power supplies, but that's all.

[Anonymous No.1009011]

Hi /diy/. Complete electronics noob here. I'm looking at the Arduino tutorials and the first one about analog input/outputs is confusing me.
I'm doing pic related. The tutorial says that changing the values of the potentiometer should brighten or dim the LED when using this piece of code:
>sensorValue = analogRead(analogInPin);
>outputValue = map(sensorValue, 0, 1023, 0, 255);
>analogWrite(analogOutPin, outputValue);
However this doesn't happen in my setup. The led is either on or off. I can get the led to brighten or dim using the following:
>sensorValue = analogRead(sensorPin);
>digitalWrite(ledPin, HIGH);
>delay(sensorValue / 100);
>digitalWrite(ledPin, LOW);
>delay((1023 - sensorValue) / 100);
But I'd like to understand why my LED doesn't react when using the first piece of code. What could I be doing wrong?

[Anonymous No.1009013]

>>998973
if you can;t start with Horowitz and work your way thru without any prior experience, then real electronics is probably not for you. just my two cents. skip all the rest and go straight to that.

That's what I did in uni, along with all of my classmates, and as shit as teachers are in most eng. classes, you basically teach yourself.

[Swifts !WiNDUPyWhk No.1009017]

>>1009011
Well, it looks correct
Did you connect the LED to pin 9 like you did in the schematic? That's the first thing that comes to mind really. Making sure the output pin is indeed a PWM pin on your Arduino.

[Anonymous No.1009027]

>>1009017
Ah, it wasn't a PWM pin! Thanks a lot.

[Anonymous No.1009034]

>>1009027
Part of the problem with arduino is that the ordering of PWM pins makes no sense at all.

The actual ATmega pins/ports used have nothing to do with the pin numbers that the arduino is labelled with.

[Anonymous No.1009047]

>>1007689

What kinds of voltages or currents are you expecting?

P = I*V = I*I*R = U*U/R

[Anonymous No.1009050]

>>1009013
Your two cents is just plain old-fashioned elitism. Not everyone wants to be a professional and even relatively basic understanding allows you to design, repair and modify useful or interesting devices. Whether they fulfill your requirements for "real electronics" is not particularly important. Also, nothing stops you from learning more later.
Sure, if the listed "beginner" and "intermediate" books seem too simplistic, there's no point in restricting yourself to those and not trying AoE or whatever the nearby university uses in their introductory courses.

[Anonymous No.1009068]

>>1009034

Arduino is effectively deprecated anyway now that Atmel sells a 10 dollar board that is fully supported with on board debugging in Atmel Studio

[Anonymous No.1009081]

>>1009068
Well, you could still buy the parts and make your own PCB and end up with it being slightly cheaper.

But if you can construct your own PCB, you probably don't need Arduino holding your hand much either.

For me, the only real advantage is the shields, allowing super quick testing with whatever modules you need without having to worry about the physical connectivity.
That's about it though, once you actually know for sure what modules you want to use, Arduino is pretty much useless.

Also, the availability of shields isn't nearly as good as people make it out to be.
Just try and find a standalone RTC that plugs into Arduino header pins without breaking them out.

[Anonymous No.1009084]

>>1009081

I'm pretty sure it's compatible with the shields

pretty much every dev board is now

[Anonymous No.1009090]

>>1009084
Well hardware-wise sure, as long as it matches the 5V (or 3.3V if it's a due shield) then it'll work.

It's just a matter of whether the power, ground, analogue input pins etc. are all in the same locations.
For reference, the Arduino often isn't even compatible with itself, the DIO locations of SPI and I2C pins are completely different between the Uno and the Mega2560.

When I say quick testing, I mean plug-and-play without even checking to see if something is the right way around (there's only one way to insert full-sized shields).
Literally retard proof.

[Anonymous No.1009113]

Not sure if I should post here or the stupid question thread.
I'm trying to connect this keyboard FFC/FPC to a breadboard or arduino but I am having trouble working out the connector I need.
It appears to be staggered pins at the end, not the same size all the way along. Does anyone know where I could find a connector for this? Thanks

[Anonymous No.1009163]

If I am absolutely new to electronics and my book requires a multimeter for some of the "experiments". Would I be fine with buying a POS multimeter like this?

https://www.amazon.com/Excel-DT9205A-Digital-Multimeter-best-DT9205A/dp/B006TOJ9Q2

[Anonymous No.1009167]

>>1009163
Check out >>1003194

[Anonymous No.1009171]

>>1009167
My apologies. Thanks.

[Anonymous No.1009177]

>>1009171
No problems, /diy/ won't get any worse and helping people is always a good thing

[Anonymous No.1009241]

So /ohm/,
>>1008671
This chart is pretty great for enthusiasts looking for their next project, however most of these projects are pretty advanced for beginners and novices, who also happen to be the ones who ask for "project ideas" the most, and may scare away some poor soul who has never even stuck an IC into a breadboard; so, what I'm proposing is we round up ideas that sound good for beginners and use that whenever someone posts something along "Hey /ohm/ what are some good projects?".
I was thinking about stuff like discrete generic ICs (741s, logic gates, 555s, simple stuff that they've probably already used and that can give an insight on their inner workings), soldering easy circuits (555 blinkers or PWM controllers, simple logic...) and the like.
TL;DR We need a noob version of this chart, any ideas?

[Anonymous No.1009259]

>>1009241
Go through the chart and filter out stuff that's too hard for noobs.

The main question is whether you want stuff that's educational or which is of practical use.

Most of the basic practical circuits boil down to choosing an IC and using the reference circuit from the data sheet.

[Anonymous No.1009262]

>>1009259
I think educational "disguised" as semi-practical would be the best approach, that's why I suggested discretes and basic soldering stuff

[Anonymous No.1009297]

>>1009241
Basic circuits is all the list would consist of.

High pass filter, low pass filter, bandpass filter, bandstop filter, 555t timer, transistor based amplifiers, op amp based amplifiers, voltage/current controlled voltage/current sources, voltage/current divider, etc

[Anonymous No.1009306]

Seems Humble Bundle is having a decent book bundle.

https://www.humblebundle.com/books/make-electronics-book-bundle

[Anonymous No.1009324]

>>1009297
Too theoretical. This is /diy/, not /sci/.

You want things which can be put in a case and used, not sub-circuits.

[Anonymous No.1009336]

>>1009324
Those things are made up of a bunch of subcircuits tho

[Anonymous No.1009340]

So, I've recently learned that I like working with electronics more than anything mechanical, especially now that I'm learning more about it. I'm looking to make a career shift to avionic technician.

What's a good beginner tool kit for someone on a tight budget? I just need the essentials, then I can pick up more tools as needed when I take on bigger projects and start going to school for what I want to do.

[Anonymous No.1009353]

what's the best way to make an electrical heater? I'm looking for something that can get a small fish tank to about 80F and run off of DC. I'm assuming I need some kind of oscillator & heat coil.

[Anonymous No.1009355]

>>1009340
I think talking electronics has a few nice kits

[Anonymous No.1009368]

>>1009353
Take your output from a wall outlet and use a full bridge rectifier to convert to DC then connect it to a relay, normally open

Connect an Arduino to the coil part of the relay and set it up as a PID controller

On the output of the relay make a coil ajd connect it from the output to the ground wire coming out of the relay


Turn it on and try not to die

[Anonymous No.1009373]

>>1009353
Buy a fish tank heater.

Ultimately, anything that conducts electricity (that isn't a superconductor) converts electricity into heat.

Designing a heating element is mainly thermal and mechanical; the electrical side is trivial (Ohm's law). The coating/casing needs to be a good thermal conductor but an electrical insulator (so that the resistance doesn't change if the element contacts itself or a conductor).

You need enough surface area and low enough thermal resistance that you'll get sufficient heat transfer (Watts) with a relatively low temperature differential (so that the actual heating element doesn't need to be so hot as to melt its surroundings).

To measure the temperature, you can rely upon the fact that the resistance of a conductor typically increases with temperature, so the heating element is its own temperature sensor.

[Anonymous No.1009408]

>>1009355
I'm just seeing parts kits. While helpful, I'm just looking for an inexpensive, but reliable kit to start me off.

[Anonymous No.1009452]

>>1009408
build a radio or servo motor controller, a power amp or supply. Built user reusable circuts

[Anonymous No.1009459]

>>1009452
My bad, I'm missing an important word in my post.

>tool kit

I am looking for tools.

My brain is still fried from work.

[Anonymous No.1009471]

>>1009324
>wants things that can be put in a case and used
>amplifiers are useless in a case
>voltage regulators and current sources are useless in a case
Excuse me you filthy swine. Pic related, a Transcranial Direct Current Stimulator (aka a constant current source for ~.5-4mA). Runs off DC or battery.

Left switch toggles the ammeter plugs open, top is the output, lower right is the battery/plug switch. On/off on the right with amp adjust in the middle.

[Anonymous No.1009475]

>>1009459
>tool kit
You can get a "PC repair kit" for like $20. I got mine years ago which had a wire stripper (black plier-shaped ones), soldering iron (ok), and a bunch of screwdrivers and pliers and wrench. The quality of them may have slipped if Amazon is any indication (got mine from newegg, a Rosewill kit). Basically here's what you need if you're doing hobbyist shit:
>screwdrivers: flat, philips, torx, usually you get big packs

large: any long-handled solid ones will do. Usually long ones are better for reaching into long holes than large screws

normal: http://www.amazon.com/Best-Sellers-Home-Improvement-Screwdriver-Bit-Sets/zgbs/hi/2445469011

small: http://www.walmart.com/ip/Ultra-Steel-Precision-Screwdriver-Set-44pc/17190613

>soldering iron
30-40W can work well. I like 40 but it can burn things to shit. If you go cheap, get a file and replacement bits. Easier to deal with shaping/replacing bits than "using it properly". I DIYed a regulator by using a plastic wall box, lamp cord, dimmer switch, and a socket (not grounded, but fine for soldering). You can get gas soldering irons.

>solder
Any lead-free solder is usually fine. I'm using "expired" solder. Desoldering braid is of differing quality and works better with hotter irons.

>multimeter
At least one. I've seen a youtube guy use four at a time to measure in/out voltage and current, but he's an actual Electrical Engineer I think.

Scissors and pliers are a must. I use a small pair of pliers (the kind that have a wire-cutting hole) all the time for some stupid reason.

[Swifts !WiNDUPyWhk No.1009479]

>>1009475
>You can get gas soldering irons
As far as gas soldering irons go, I can vouch for these cheap ones.
What's great about it is that you can use Weller tips with them.
The stock tips will make you want to kill yourself, but once you replace the tip it becomes pretty much equal to a Weller Pyropen.

The soldering iron itself cost 25€ with the Weller tip going for 16€.

[Anonymous No.1009481]

>>1009475
I figure this might serve me well?

https://www.amazon.com/Rosewill-Professional-Computer-Components-RTK-090/dp/B004H07POS/ref=lp_13825561_1_10?s=pc&ie=UTF8&qid=1466060156&sr=1-10

This is another thing I'm liking about this, it seems like I don't need a lot of tools unless I'm going into a specialty field.

[Anonymous No.1009515]

>>1009471
> amplifiers are useless in a case
An amplifier is useful. But a practical amplifier is basically connecting the input, output and power jacks to the pins on a chip.

Building an amplifier from discrete transistors is either an exercise rather than a practical device, or (if you're talking about something that's actually an improvement over a single-chip amp) far beyond beginner level.

On the "op amp based amplifiers" front, you could probably make some useful pre-amps, but they're fairly specialised (not everyone plays guitar) and tend to be "here's a circuit, build it". So it's either just a soldering project (if you don't attempt to understand it) or an academic "signal processing and op-amps 101" exercise (if you do).

Ideally, we want things which a) are practical, b) can be used to gain some understanding of circuit design, but c) don't feel like they were pulled from the exercises section of a textbook on linear systems theory.

> voltage regulators and current sources are useless in a case
Neither of those were in the list. What was in the list was:
> voltage/current controlled voltage/current sources
Those are sub-circuits, not devices.

[Anonymous No.1009525]

>>999459
Use a really big capacitor, esp8266 only uses power when its transmitting.

[Anonymous No.1009529]

>>1002209

Just fookin bend it.

[Anonymous No.1009750]

I'm a fist time Multisim user and I'm in a bit of a bind. I'm designing a circuit around a MC68HC11 microcontroller. I have the academic version of Multisim and it doesn't have that particular chip in the parts library. I know the professional version has it. Is there anyway to get that part without having to buy the professional version? Second, when I export that part (using the 7 day trial version of Multisim Pro) to Ultiboard it includes only the footprint for the chip disregarding the fact that it's supposed to go inside a PLCC52 socket with an entirely different footprint. Is there any way to fix this? Any advice would be appreciated.

[Anonymous No.1009823]

>>1009750
>without having to buy the professional version?
Be a filthy pirate or use someone else's (school's?) professional version.
>footprint
Never used Multisim, but normally you'd either change the footprint from schematic, change the footprint in library, make another component with the correct footprint or use another very similar component (in correct package) and change the name in schematic.
Note that if you draw the schematic and/or the pcb symbol by yourself, then there's no need to limit yourself to Multisim. Well, unless using Multisim + 68HC11 is some weird course requirement and you absolutely need to simulate the circuit with processor and all. In that case school should provide you the required tools.

I've been wondering this before, but why is adding components and footprints to libraries often seen as a nearly impossible task?

[Anonymous No.1009829]

>>1009823
I've tried making a custom part with the footprint I need but while ultiboard has at least 3 acceptable footprints in the IC socket category, multisim doesn't seem to recognize anything from the socket category. I should probably add that I won't actually be simulating anything. I'm just drawing out the schematic so I can send it to ultiboard. Am I going about it the wrong way?

Also this is an independent study project for school so I'm mostly on my own when I run into problems.

[Anonymous No.1009919]

>>1009829
I don't know if Multisim has different workflow than the most other programs, but in any case, creating new symbols is one of the most basic and most important things, so there's bound to be tutorials about the topic.
Maybe someone else knows what the problem is exactly, but as a guess: the schematic symbol pin numbers must match the pcb symbol pad numbers. Maybe your PLCC socket symbol uses PGA-like numbering (A1,A2,...), while the schematic symbol uses PLCC's numbering (1,2,...).

[Anonymous No.1009988]

Dumb question

See the picture and tell me if i can put the diode across the 2 wires further away from the actual relay, the 2 stars are where i would attach the diode. OR does the diode need to be at the contacts as close as possible? Dumb question but please answer.

[Anonymous No.1010030]

>>1009988

it's good to reduce induction by placing components close to each other, but it will almost surely be fine since that diode is probably meant to filter out inductive spikes from the relay's coil anyway

[Anonymous No.1010040]

>>1009988

the diode is to protect the thing at the opposite end of the wire, so you can place anywhere along the cable, and it'll work the same.

[Anonymous No.1010046]

>>1010040
>>1010030
thanks guys! just making sure.

[Anonymous No.1010142]

If this belongs in /sqt/ then with your command I will delete this post. If I am an absolute beginner to electronics. how should I go about getting into it? Should I use the Art of Electronics as a reference book or should I read through the entire book? I was thinking about reading through Make:Electronics with The Art of Electronics as a supplemental book, if this is a stupid idea please let me know. Thank you.

[Anonymous No.1010173]

>>1010142

nah, dude, you dont wanna start with any heavy duty theory crap. you wanna get the series of booklets Forrest M Mims did for Radio Shack. get a few components from an ebay kit, along with a small breadboard, build the circuits, and play around with modifications. if you're still interested by then, you can move from there in one of several directions:
- read the theoretical stuff, learn formulas, figure out all the stuff works, or
- get arduino'ed, buy shields, learn to code.

(the Mims booklets can be found in torrents)

[Anonymous No.1010187]

>>1010173
OTOH, if you want to do anything other than build circuits from complete schematics, you're going to need at least some theory.

The main concepts to understand are voltage, current, charge, resistance, power and energy. Kirchhoff's laws are useful but you can mostly get by with understanding series and parallel.

Once you start dealing with capacitance and inductance you need basic calculus.

[Anonymous No.1010348]

>>1001182
Look up the datasheet for that op amp. Two of those open pins are the positive and negative power input things.

[Anonymous No.1010393]

>>1007031
Fellow autistic EE here, your question shows that you areat the beginning of a long and bumpy road.

1. She is probably having suicidal thoughts on the reg for not being pretty enough to be a trophy wife or get 1000$ per instagram posts and instead having to wage slave in a career which probably can't fulfill most of her needs.

2. This means that probably the last thing she wants to talk about is nerdy EE shit in a "dating" context

3. In general your knowledge of Nixie tubes won't get her wet sadly, things are a bit different to compared to male-bonding

4. You have two choices to get her interested: -being rich or being in the genetical elite. I hope you are sorta rich via being a probably top tier EE nerd but rich men are only good for getting cucked and getting financially raped in a divorce.

5. The only way to get her honestly interested is being good looking enough. That sorta goes like FACE > Height > Frame > other shit. FACE. FACE. FACE. It would be hard to get all the important stuff into one post but you should have a prominent jaw line, wide palate, medium/short philtrum, not overly deformed nose, NW0-NW1 hairline with an up-to-date haircut, a good FWR and a top-tier eye area. The latter means preferably having a positive canthal tilt, low to no scleral show, preferably no upper eyelidd exposure, having good orbital bones etc.
To sum it up EE girls like the same type just like any other girls, they are human after all.

[Anonymous No.1010402]

Hello /ohm/.
Amateur mechatronics engineer here.
Today I has released a video on Youtube.
>> Youtube URL : https://www.youtube.com/watch?v=5M5ao1D4jWo
450kV Audio modulated Tesla Coil & Arduino controlled robot orchestra cover version of song named "Hello" by OMFG.
It took about 2 month to make the whole project.

It's little bit geeky, but I'd like to hear your opinion about the project.
I'm sure that there will be better electronic engineer than me around here.
Please mention if you have suggestions or improvements, etc.

[Anonymous No.1010411]

>>1010402
>>1005237
>>1010393
Oh, and speaking of NIXIE tubes, there are some nixie tube & numitron tube porn in front of the video.(I highly recommend you to see the video if you are nixie tube lover)

and if you visit my channel, you can easily find my IN-12B tube nixie clock making video...
I hope this will help(if you're trying to make one of those).

[Anonymous No.1010694]

what are typical ESLs for the various types of capacitors? they don't seem to be published anywhere and i'm too lazy to reverse engineer an impedance chart

[Anonymous No.1010697]

>>1010402

Needs more bass. Whole percussion line needs work, actually.

But that's mostly nitpicking and, if you got this far, not too hard to remedy.

Build itself is 4/5 bretty gud.

[Anonymous No.1010706]

>>1010402
impressive.

[Anonymous No.1010712]

>>1010694
>can't read off a straight line on a log plot
ok breh

[Anonymous No.1010746]

>>1009479
How long do these usually take to get warm enough? I have a weller one laying around which kinda disappointed me with how long it takes.

[Anonymous No.1010766]

>>1010694
Typically a few nH, varies with physical size and capacitance.

Some data sheets will provide a formula or graph for ESL, some for resonant frequency (which is 1/(2*pi*sqrt(L*C))).

If it's not specified (which is fairly common for electrolytics other than pulse-discharge types), it's because you wouldn't use the capacitor at frequencies high enough for ESL to matter (the ESR will become an issue long before the ESL does).

[Anonymous No.1010912]

Guys what's the average temperature for 100W 4R power resistor? picture related

[Swifts !WiNDUPyWhk No.1010917]

>>1010746
Well it depends on the tip.
The standard conical one takes around 20 seconds, a chisel tip takes slightly longer since there's a bit more material to heat.

All in all, quicker than any electric iron I've ever used.
Also you can use standard cigarette lighter butane. The Weller my dad used to have was really picky about the gas and only worked properly with their own brand which cost stupid amounts of money.

[Anonymous No.1010921]

>>1010912
You should check the datasheet, but generally you can expect that it can take 100W only at 25C ambient temperature and when mounted in a suitable heat sink.
It will get very hot when you run it at 100W. No idea exactly how hot, but 200C wouldn't be surprising.

[Anonymous No.1010933]

>>1010921
no datasheet available, i mounted them to 2mm thick 40cm long sheet of galvanized steel. still waiting for more supplies before i can run a test. 40cm is because i plan to add 1200W worth of power resistors

i expect to run them at 70W each

[Anonymous No.1010943]

>>1010933
They're almost guaranteed to overheat.
If there really aren't any useful markings in them for finding the correct datasheet, you could check Vishay's RH series to get a rough idea.

[Anonymous No.1010950]

>>1010912
> what's the average temperature for 100W 4R power resistor?
Meaningless non-question.

>>1010933
> no datasheet available,
In which case, they're of no use. Anything you do with them is an experiment to determine the MTBF under those operating conditions.

[Anonymous No.1010955]

>>1010943
TYPO!

I meant 40W, I forgot 2 resistors in series changes the power rating of them. not sure where i got 70W from. It will be 6 lots of 2 resistors in series so its operating under 50% of their rated 100W

>>1010950
I guess the worst that can happen is they blow a fuse if one fails..

[Anonymous No.1010968]

>>1010955
> I guess the worst that can happen is they blow a fuse if one fails..
Not even that; resistors always fail open-circuit, never short-circuit.

The worst that's likely to happen is that you waste time and money on something which won't work.

You need two things from the data sheet: the element-to-case thermal resistance, and the maximum element temperature. Those will let you determine the maximum case temperature at a given power dissipation.

Then it's just a matter of calculating (or measuring) the thermal resistance of your heatsink to determine the actual case temperature at a given power dissipation. From that, and the maximum ambient temperature, you get the maximum power dissipation.

But you can measure the properties of the heatsink. You can't measure the element-to-case thermal resistance unless you can measure the element temperature, and determining the maximum temperature would require sacrificing a statistically-significant number of samples in experiments whose duration is equal to the required life expectancy.

It's far easier for the manufacturer, as they understand the failure mechanisms and can reliably extrapolate from much shorter tests.

[Anonymous No.1010983]

>>1010955

also consider doing something like this especially If you are making a dummy load.

https://www.youtube.com/watch?v=cqwUjA8vgKk

you can run them way above the power rating if you put them in a tank like this, or just get the resistance you need from suitable ohm wire and submerge them like he does in the video

[Anonymous No.1010984]

>>1010983

fucking hell wrong link

https://www.youtube.com/watch?v=WECW88rJYrE&spfreload=10

[Anonymous No.1010986]

>>1010984
lol thought you were trolling for a minute

that does look like a better idea than power resistors.

[Anonymous No.1011016]

>>1010411

I was the guy in the other thread that asked you about the 8 segment nixies. I'm pretty sure I know the layout that I want, but I may check it out to get some more ideas. Love the setup you got going by the way.

[Anonymous No.1011503]

>>998973

So I just switched to linux and I need a nice supported embedded debugger that just werks. Previously I have used cheep ISP adapters and the Atmel Xplained Mini.

What is the best (most modern) Pic supported by the Pickit 3?

also what the fuck is a "processor pak" that it talks about in the Pickit 3 manual?

[Anonymous No.1011506]

>>1001618

is it just chink shit or actually an officially supported board? The only one I can find on their website is almost 100 bucks