/diy/ - Do It Yourself

[/ohm/ No.799233]

Old thread:
>>794221
>I'm new to electronics, where do I get started?

There are several good books that are commonly recommended for beginners and those wanting to learn more. There are also plenty of good Youtube channels that teach about the basics as well as advanced concepts. The best way to get involved in electronics is just to make stuff. Don't be afraid to get your hands dirty. Take something apart or build something you find cool on one of the many electronics websites.

>What books are there?

Beginner:

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

Intermediate:

All New Electronics Self-Teaching Guide by Harry Kybett, Earl Boysen
Practical Electronics for Inventors by Paul Scherz and Simon Monk

Advanced:

The Art of Electronics by Paul Horowitz

>What Youtube channels are there?

https://www.youtube.com/user/EEVblog
https://www.youtube.com/channel/UChturLXwYxwTOf_5krs0qvA
https://www.youtube.com/channel/UCosnWgi3eorc1klEQ8pIgJQ
https://www.youtube.com/channel/UChtY6O8Ahw2cz05PS2GhUbg
https://www.youtube.com/channel/UC1rxMIOt82ieNE19w15U5YQ

>What websites feature electronics projects? Where can I get ideas for projects?

https://www.adafruit.com/
http://www.instructables.com/tag/type-id/category-technology/
http://makezine.com/category/electronics/

>Where do I get components and lab equipment from?

http://www.jameco.com/
https://www.sparkfun.com/
http://www.ramseyelectronics.com/
http://www.allelectronics.com/
http://futurlec.com/
http://www.ladyada.net/library/procure/hobbyist.html
http://www.mouser.com
http://www.alliedelec.com
http://www.newark.com
And sometimes even just ebay.

>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.799235]

Would this circuit work as a power supply short circuit protection?

[Anonymous No.799237]

>>799233

Hey h/ohm/os, I've a question. I know that for more advanced study of electronics, you need higher level math. I want to know what level of math these books on electronics assume of their readers, and whether someone with only a calc 1 and no physics background would fare well reading them.

[Anonymous No.799254]

>>799235
Wrong pic?

[Anonymous No.799257]

>>799235

nope, that circuit doesnt even make any sense. you have an op-amp without feedback, so it acts as a comparator. you're comparing an input voltage against ground. against ground, FFS! so the op-amp accomplishes nothing, you might as well eliminate it and drive the transistor base directly. then you have a PNP switch (turns on when base is grounded) that doesnt seem to accomplish anything.

here's what you should be trying to accomplish instead: use a tiny resistor, say 0.1ohm, in series with the power supply's negative terminal. say you wanna limit the current to 3A, then you need a comparator that will measure the drop across the resistor, and if it goes over 0.3V, you shut down whatever element is passing the output current.

[Anonymous No.799260]

>>799237
Afaik none of the books mentioned in OP is math-heavy. However, instead of relying on /diy/'s opinions, you should check the books and see if they're suitable for you.

[Anonymous No.799268]

>>799237
I'm studying EE and the higher math level required to do well on electronic related subjects is differential calculus and basic complex analysis. Most of the time you'll be using plain formulas and relationships from high school though, specially at the beginning.

[Anonymous No.799278]

>>799260

Oh, I'm just loaded on units this semester and wanted something to fill any spare time without having to drop the book part of the way through because of the math. Thanks for the input.

>>799268
Thanks, I'm actually planning on transferring into a 4 year school from a CC to study EE as well, so your info is also helpful. I figured I'd start reading up on major specific stuff before it comes time to take the classes. Good luck to you.

[Anonymous No.799291]

>>799257
Could you help me with this circuit? Lets say I have a 0~30V 6A power supply and I want to build a short circuit protection using an op-amp, how should this circuit be?

[Anonymous No.799310]

>>799291
You need to put a high wattage resistor with a very small value (a fraction of an ohm) in series with the output, then, set the OP amp as a comparator. The current flowing through the resistor will create a voltage drop that can be compared with an OP amp. If the voltage exceeds a treshold the OP amp turns on and activates a latch, stopping the circuit.

You'll need to do some calculations, to calculate this voltage drop, multiply the maximum current per the value of the resistor set the OP amp to trigger at that voltage level.

This might be inconvenient if you plan to work with high voltages or inductive circuits, since voltage spikes might cause false triggers or destroy the IC.

[Anonymous No.799316]

>>799291
You need to put a high wattage resistor with a very small value (a fraction of an ohm) in series with the output, then, set the OP amp as a comparator. The current flowing through the resistor will create a voltage drop that can be compared with an OP amp. If the voltage exceeds a treshold the OP amp turns on and activates a latch, stopping the circuit.

You'll need to do some calculations, to calculate this voltage drop, multiply the maximum current per the value of the resistor set the OP amp to trigger at that voltage level.

This might be inconvenient if you plan to work with high voltages or inductive circuits since voltage spikes might cause false triggers.

Not related to the thing above: I'm making a plasma speaker and my IRF250 mosfet get's too hot, can I put two in parallel? Can someone recommend me a capacitor that can handle more power?

That mosfet is switching a 160 watts load with 4 amps.

[Anonymous No.799324]

>>799316
You can parallel mosfets. You don't even need balancing resistors.
If you can't find a reasonably priced big enough and good enough capacitor, put several smaller in parallel.

[Anonymous No.799344]

>>799233

What's /ohm/'s opinion on /sci/'s electronics engineering books recommendations:

http://4chan-science.wikia.com/wiki/Electrical_and_Electronics_Engineering

[Anonymous No.799345]

>>799233

What's /ohm/'s opinion on /sci/'s electronics engineering book recommendations:

http://4chan-science.wikia.com/wiki/Electrical_and_Electronics_Engineering

[Anonymous No.799385]

>>799324
Sorry I just meant mosfets, not capacitor. Good to know I can parallel mosfets though, thanks!

[Anonymous No.799392]

Hey, I'm using atmega16 for learning and shit so I was wandering will this PWM work?
here's the C code:

#include <avr\io.h>
#include <avr\iom16.h>
#define OCP PIND5

int main(void){
DDRD=(1<<OCP);
TCCR1A= (1<<COM1A1)|(1<<COM1A0)|(1<<WGM11);
TCCR1B= (1<<WGM13)|(1<<WGM12)|(1<<CS12)|(1<<CS10);
// values to ICR1 and OCR1A here
for(;;){}
}
Now I know ICR1 and OCR1A are missing but assuming they are there would it give pulses if I hooked it up to oscilloscope?

Help much appreciated.

[Anonymous No.799404]

>>799345
Most of that stuff seemed to be widely used university level textbooks. If you like the university way, they're decent.

[Anonymous No.799412]

Hi guys.
As I finished my nixie clock that you can see here
>>799363
As I am a stoopid boy I simply bought the kit from pvelectronics.
But I have a question about what the hell I've done. In the picture related you can see the first part of the circuit.
J1 is the "PCB power socket" (the femal plug)
And the fuse right after J1 says:
"500mA so don't put more than 500mA in me or I'm gonna shit your shit"

Am I right ?

[Anonymous No.799418]

>>799412
I will go on with thinking that I'm right.
So the instructions of pvelec' says:
"Output 12V DC regulated, minimum power output capability of 500 mA "
My output is 600mA. Everything works but seriously why don't my shit get shited?
Picture related, my shit without shit.

[Anonymous No.799421]

>>799418
doesn't it say minimum 500 and aren't you giving over that?

[Anonymous No.799422]

>>799418
Typical electronic gadgets draw only as much current as they need. It is irrelevant how much more the power supply can deliver, the extra current is not used. Thus, your fuse will not blow even if you used a 100A power supply as long as the voltage is correct and your clock is operating correctly.

[Anonymous No.799423]

>>799421
The instruction says so.
But the fuse is 500mA. Giving it over that doesn't make it blow?
If the 500 mark is a minimum, when does my fuse blow?

[Anonymous No.799425]

>>799422
So that's not a security, it is only a filter to give the circuit the 500mA it need and no more ?

[Anonymous No.799430]

>>799423
>>799425
I think so, there is a maximum current it can conduct but it isn't labeled as "minimum power..."

[Anonymous No.799436]

>>799425
The fuse would protect against a surge in voltage.

I don't understand what you mean by "filter". Fuses protect the circuit from being overloaded by failing before anything else does.

[Anonymous No.799448]

>>799418
fuses work by blowing up before other stuff do. so the 500mA blows if your system starts pulling more than that.

[Anonymous No.799452]

>>799436
I thought that the fuse would protect against a surge in amperage.
The word filter is a bad choice. funnel is better.
>>Fuses protect the circuit from being overloaded by failing before anything else does.
That's what I thought but I was thinking that the limit was the 500mA indication.
If I had 24V and 500mA would it blow?

[Anonymous No.799455]

>>799345
Pretty much what >>799404 said. They're all university level books, the ones I put in the OP I tried to make a little more beginner friendly.

[Anonymous No.799457]

>>799448
ok I think I understood
SO as long as the circuit pulls the 500mA everything is fine, and if a short cirucuit or any other bad stuff drain more it all blow.
That's why >>799422 said that I could give 100A.
OOOOk. That is working just like the fuse of my house. Thank you all for helping me to understand this,

[Anonymous No.799460]

>>799452
Voltage is related to current though Ohm's law, V=I*R. Voltage sources can't spontaneously provide more current than what the load (resistance) is drawing, because then it would violate this law.

In a DC circuit, resistance isn't going to change much. So by the formula, if voltage doubles then so does the current. The fuse protects the circuit against voltage surges and idiots who use the wrong voltage.

[Anonymous No.799461]

>>799233
>/ohm/ in the name
>Not in the subject or OP
Ya dun goofed

[Anonymous No.799603]

Where can I chew through some online texts to learn how to make my own clock with non-standard multi-segment LED display?
Someone in the last thread said:
>>798270
>>Is 8 the highest number of outputs one can have?
>No, byte-wide shift registers are just common. You can also get shift register -like LED driver chips which can handle dimming as well.
but I was looking at RGB LED shift registers and I can't tell how they work. I'd like to read about it instead of asking dozens of questions to people in person repeatedly until I get an answer, but I'm terrible at research. I tried to search for those RGB LED shift registers, but I couldn't figure out how they work, or how to find one for my needs.

[Anonymous No.799617]

Question. Does anyone happen to know if there's a specific name for this transformer and know a place to buy it? I can't seem to find the datasheet on it and all I got to go by is 8011 7409A written on the side.

[Anonymous No.799654]

>>799603
What can't you figure out about them? An RGB LED is three LEDs in one package (often with all cathodes or all anodes connected together). Drive them with three outputs from the shift register. The common anode or common cathode means you need to pay attention to if your shift register can sink or source current.

You're having trouble doing research because there aren't really any shift registers specifically designed for RGB LEDs. A shift register with eighteen outputs can drive six RGB LEDs or eighteen individual LEDs. There's no important difference.

[Anonymous No.799665]

>>799617
I assume that's a flyback transformer for a backlight. If the numbers don't come up in google, you're unlikely to find an exact match.

[Anonymous No.799673]

>>799654
Well, can't you use a shift register to control the RGB values with fewer wires that are duplicated across all the LEDs? Basically have the RGB values coming out one, and the other being a simple on/off switch? or is that what the shift register is for to begin with? Because I'm looking at a minimum of 12 separate bars per digit.

No, I just suck at research. It's universal over all the subjects I've studied over the decades. After my initial enthusiasm about this project {this isn't the first time I've taken it on. I've had this for ages.} I've lost steam.

[Anonymous No.799683]

>>799673
It sounds like you're referring to multiplexing. Well, it's one way of doing things. Compared to the basic one output per LED arrangement, you need conceptually slightly more complex hardware (but requiring less components) and more complex software to run it.
http://en.wikipedia.org/wiki/Multiplexed_display

Shift register is just one way to "convert" few digital signals to multiple signals or vice versa. They aren't directly related to multiplexed displays or required to implement them.

More than anything, it sounds like you're underestimating the amount of research you need to do to and the level you need to start from.

[Anonymous No.799688]

>>799683
I don't think I need multiplexing to do this. I don't know if that's something I'd want to do or not.
Can't the RGB values come out of 3 pins and lead to like all the LEDs of two digits, with all 24+ LEDs being controlled by other pins? Can you not have the LEDs connected only to a shift register?
They also told me about I2C, which sounds better than the normal shift registers I was seeing.

>More than anything, it sounds like you're underestimating the amount of research you need to do to and the level you need to start from.
I have no idea how you "research harder". I just get discouraged/angry after a certain point.

I took robotics for a semester. Did well with microprocessor {probably due to lots of coding experience}, but not so great with the electronics.

[Anonymous No.799748]

>>799688
While searching for the full name of I2C {"Inter-Integrated Circuit"}, I found this "sparkfun" site. Maybe it will be helpful enough to teach me everything I need, since it also lists suggested reading topics which are links to more of their own lessons.
https://learn.sparkfun.com/tutorials/i2c

[Anonymous No.799762]

>>799688
>I don't think I need multiplexing to do this
You don't. It's just one, quite popular, way to drive displays. Your explanation sounded like you were to use it.

>They also told me about I2C, which sounds better than the normal shift registers I was seeing.
It's better in the sense that you need only two processor pins for it. It's worse in the sense that it's slower, more difficult to program and the number of similar devices per bus is often quite limited.
My suggestion would be to find a LED driver you like (number of outputs, features, package, supply voltage, whatever) and live with whatever interface it has.

>I have no idea how you "research harder".
It sounds like you're having difficulties understanding datasheets, shift registers and LED drivers, among the others. You'd need to read more about all those. If that's not going well, then maybe you'd need to start from more basic stuff, like from some introductory electronics textbook.
Another way would be to start by copying someone else's clock design and then modifying it bit by bit to the direction you want and doing research whenever you run into problems.

[Anonymous No.799798]

Posting challenges again.
Sorry for the typos, I still haven't fixed it yet.

[Anonymous No.799806]

I have a question about electronics that is so fundamental and basic that it becomes difficult to know how to ask it properly but I will try.

How do people who do electronics know that they must use a certain set of components to achieve something they want to do? I mean how can you learn how to achieve the best solution to a problem if there are so many ways of solving each problem? What is the process of deciding which components to use?

[Anonymous No.799808]

>>799806
It's mostly knowing exactly what every component does. Like if I'm working with mains electricity and need to step it down to 24V or something, the obvious choice would be a transformer, and then if I need to convert that to DC voltage, there's only a few ways to do that.

> I mean how can you learn how to achieve the best solution to a problem if there are so many ways of solving each problem?
There's never really a best solution (unless it's a really simple circuit). If ultra-low-power is a goal of your circuit, it will probably look different from someone whose goal is an easy-to-manufacture circuit, or someone who is tight on budget and already owns a component that can be made to work like another component that would be more expensive to purchase.

I guess your best bet is practice designing circuits, reading tutorials, and, well, just learning about electronics mostly.

[Anonymous No.799815]

>>799806
It's intuitive, I suppose. You spend a lot of time learning about various components and then when you have a tangible design idea, you already know more or less what you'll need to implement it.

---

Just a few stupid questions of my own, /diy/:

1) What's the worst that could *really* happen if I control a 10 A universal motor with a TRIAC but exclude a snubber? I've heard about "snubberless" TRIACs. Are these basically the ubquitious standard nowadays even if the word "snubberless" isn't explicitly written on the datasheet, or do I need to actively shop for one?

2) If I accidentally put a 700 mA load (a small incandescent lightbulb) on an Arduino Uno's digital pins (max. 40 mA) without a current-limiting resistor in series with it, and the bulb refused to light up, is there any chance I've done permanent damage to the board?

3) Should I bother adding a bleeder resistor in parallel with the DC pins of a bridge rectifier if the whole thing is going to be sealed inside a package anyway where no one will ever be able to touch it?

[THE GOD MACHINE No.799820]

>>799798
What you expect to happen when you finished them all :
Get accepted into the real /ohm/ crowd, free parts and secret channels will open for you
What actually happens :
Your house will be full of useless shit

[Anonymous No.799851]

>>799798
Are these supposed to be analog only?
Whew

[Anonymous No.799855]

>>799815
1. Can't say I'm that familiar with TRIACs, but afaik your control will just work badly. And the old TRIACs didn't go anywhere.
2. ATMega328's outputs aren't short circuit protected, so it should be possible to kill them by overloading, but at least they don't seem to die immediately. If the pin still works and the processor stays cool, I'd assume the pin is ok.
3. It's a good design practice and devices which stay half-alive a long time after power off are obnoxious.

>>799851
No, there are projects like "digital clock".
>>799798
Why are things like "vacuum chamber" on that list?

[Anonymous No.799870]

I need something that can act as a small axle and measure how fast it is being rotated.

I know a motor like pic related can do it, but is there a device specifically designed for this purpose?

[Anonymous No.799895]

>>799870
You mean a techometer? They usually use a magnet and hall sensor or photogate to output pulses at the rotation period.

[Anonymous No.799907]

I have this idea of making about 5 LEDs light up from the middle to the edges similar to the top of this gun here
https://m.youtube.com/watch?v=PPV7CJoQW8Q
I think some transistors and capacitors can do what I need. However I dont know how to make them light up as fast as I want them to. Is there an equation for this or should I just try and guess?

[Anonymous No.799908]

>>799392

just try it yourself. Not familiar with that chip, but PWM should either be accomplished through set up or at least interrupt driven

[Anonymous No.799910]

>>799907
just to make a light flash you need something like an astable multivibrator which is just caps trannies and resistors. you vary the on/off time by changing the cap values or the resistor values (caps charge via resistors see?)
or you could use a 555 timer but there are limitations and you still need lots of external components.
if you want some larson scanner type thing its probably easiest to use a decade counter driven by a 555, or a shift register i suppose if you are a masochist.
to get the full effects of HF pulsing while moving around your best bet is going full embedded. Atmel AVR 8-bit (with enough output pins - or shift register remember?) will probably be your cheapest overall considering programming costs if you are willing to go 100% on this.

[Anonymous No.799914]

>>799910
Shit. There goes that idea then

[Anonymous No.799915]

>>799806
usually what happens is you open stuff up and figure out how it works, or look at schematics other people post on the internet (or hobby magazines if you are old)
after a while you get to know the various 'blocks' you need to fit together to complete some task.
It's not some skill people learn to look at a circuit and know exactly whats happening immediately, usually circuits are drawn to be recognizable, so someone can pick out what each part of the circuit does and then work out the function. if its badly drawn or you don't recognize the parts then you have to work it out.
if you are designing a circuit then again its like designing anything, you break it down into sensible, logical and manageable steps until you can implement it with components.
then you draw them sensibly, someone builds it, it doesn't work and you are back to square 1.

[Anonymous No.799919]

>>799914
problem?
you can get a larson scanner soldering kit from most places, not close enough?

[Anonymous No.799930]

>>799919
The larson scanners pcb is too big and it doesn't have the same effect I need. I guess I'll have to invest in a debugger then.

[Anonymous No.800058]

>>799855
>Why are things like "vacuum chamber" on that list?
There's a lot of electronic controls that go into creating a vacuum chamber.
Also, I started running out of ideas.

[Anonymous No.800070]

So, /ohm/

Decided to take on the Home Made Power Supply project.

And now I'm at an impasse.

Do I go dual-rail adjustable with 2 LM317's?

Or do I go tri-rail fixed with a 7805, 7833, and 7812?

I'm mostly an embedded microcontroller guy, so I will largely only need 3.3V and 5V for awhile, but planning for the future I can see the argument where testing my circuits for low-voltage situations would be good.

What say you, /ohm/? Convince me one way or the other.

Pic related, it's the bare beginnings of the design, which I may scrap.

[Anonymous No.800071]

>>800070
The way I built mine, I have some DPDT switches on the adjustment resistors. I can select between a potentiometer, 3.3V, 5V, and 12V straight from the wall wart I used as an input. It's only single rail, but it wouldn't be hard to do it with a negative regulator too. It's been really useful. Probably not the most accurate, and you have to calibrate the trimmers, but it's good enough. Also do yourself a favor and include an LED and a constant current source.

[Anonymous No.800077]

>>800070
put the switch on the other side of the transformer.

[Anonymous No.800087]

>>800070
It's easy enough to build a new one if you need a different set of features.

[Anonymous No.800090]

>>800070
There are plenty of things which run at some other voltage than 3.3V or 5V and many things which ultimately use one of those voltages take unregulated input of some other voltage. You might also want to make devices with unregulated supplies (like 2 x AA).
Adjustable power supplies are also useful for testing things like sensor inputs or whatever analog stuff your devices have.
Or you might want to power something made by someone else once in a while.

I think one adjustable output would significantly increase the usefulness of the supply, even if you have 3.3/5/12V fixed outputs.

[Anonymous No.800095]

>>799907
What you're describing is called an LED chaser circuit. If you google that you'll find circuits that make a single row of LEDs sweep which one is on left to right. Most will be capable of driving two LEDs in parallel or series from the output, so create two rows running from the center in opposite directions.

[Anonymous No.800104]

>>800070
I made this adjustable PSU with two LM317 for the outputs and a laptop charger. Both outputs can go from 1.25 to 16 volts, an the maximum amperage is 1.5 or 2 amps in peaks, whie the 78xx can only output 1 amp.
I also must say the LM317 are heavy duty stuff, I've been abusing the LM317s by drawing more current, driving flybacks with discharges that have killed mosfets, transistors and regulators and doing all sorts of crazy things with it, and it still works.

If you want more than 1.5 amps then you can use an lm338 (5 amps) or use a lm317 in combination with a power tansistor (e.g 3055)

Here's my instructable about the PSU:
http://www.instructables.com/id/Fully-adjustable-double-output-power-supply/

>>799907
http://www.instructables.com/id/Fully-adjustable-double-output-power-supply/
A "knight rider" circuit will do, just connect another strip of LEDs to the already existent so they light up in opposite directions and stop at the center.

[Anonymous No.800108]

>>800104
Cool instructables you got there man. Nice to see "instructables" visiting /ohm/ hehehe

[Anonymous No.800117]

>>800104
psu on breadboard?
not very smart

[Anonymous No.800118]

>>800117
>psu on breadboard?
>not very smart

Not him, but why not? He's clearly prototyping it. As long as he doesn't exceed the rated voltage & current his breadboard, what precisely is the problem?

[Anonymous No.800129]

>>800118
Poopy contacts and parasitic inductances and capacitances.
Probably not problems with that particular PSU, though.

[Anonymous No.800146]

>>799233
>Not listing digikey.com for getting parts

[Anonymous No.800152]

>>800108
Glad you like it, many projects came out of boredom and they're not too complex. I've always liked hackaday more, but my projects are still too simple to be posted there.
>>800129
If you leave enough space between high voltage contacts it won't spark under the breadboard. I always considered the inductances and capacitances of a breadboard negligible.

[Anonymous No.800212]

>>800152
Switching regulators can have currents changing 100 amps per microsecond or more. That would cause a voltage drop of one volt in a 10nH inductor and a short piece of wire has can have more inductance than that.

[Anonymous No.800217]

>>800146
Every thread someone complains about this.
Every thread we mention the 2000 character limit.
Every thread we say we'll put them in pastebins so we can add more resources.
Every thread we don't do it.


Is this a new meme?

[Anonymous No.800218]

>>799233
hey /ohm/ what's a free easy chunk of software I can use to draw schematics tried of drawing by hand or using MS-Paint

[Anonymous No.800225]

>>800218
>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

Please read the OP.

[Anonymous No.800229]

>>800212
Calm down, dude. Its a bench power supply for testing small things and not a high precision nasa military grade atomic rocket controller power supply.

[Anonymous No.800268]

>>800229
It applies to many common switchers, including ones which might interest hobbyists.

[Anonymous No.800337]

Hey /ohm/

What am I supposed to do with the Vref pin of a TDA1543 DAC? I was under the impression that it should be connected to 5V but the datasheet says it's an output pin, so I guess it shouldn't.
Either way, I can't seem to make it output any other voltage than about 1.1V. Not sure if it's due to an incorrect circuit or an error in my attempt to bit bang i2s.

Datasheet: http://www.docethifi.com/TDA1543_.PDF
Code: http://pastebin.com/aSRPPaAF

Greatly appreciate any help!

[Anonymous No.800339]

>>800268
Well, thanks for pointing it out then, I'll look into it and see what can be done.

[Anonymous No.800351]

What's a good PCB-printing service? For USA at least.

[Anonymous No.800365]

>>799748
Huh. I should have read more carefully. It just teaches you the protocol. I still don't know what products are available or how they work like with RGB LEDs.
I don't know what I'm doing... I wish I had the energy to overcome this disappointment.

[Anonymous No.800368]

> What's a good PCB-printing service? For USA at least.

http://pcbshopper.com/

[Anonymous No.800374]

>>800365
I2C is more complicated than SPI. You can't use I2C for generic shift register interfaces; only ICs designed to to accept I2C.

SPI is just the regular shift register interface where you have a clock and data wire. Send a byte by changing the data line to each bit with a clock pulse between each change. An SPI port is just asynchronous hardware that will do this for you instead of an implementation in code (bit banging).

For shift registers I have two suggestions:
74HC164. This is a standard eight bit shift register. It doesn't run above 7v so you will need something like a ULN2803 or row of transistors to drive 24v LEDs. You will also probably need current limiting resistors because you're working with voltage source outputs.

WS2803. This is an IC specifically designed for running LEDs with all the bells and whistles. It has eighteen outputs (enough for six RGB LEDs). The logic power supply is separated from the output supply so it can drive up to 30v 30mA LEDs with no additional hardware. It uses current outputs so you can connect it straight to LEDs with no resistor. It behaves like a shift register, but you push in 144 bits (eight bits per each of the eighteen outputs). This is because it has internal PWM dimming capabilities; LEDs can be at 256 different brightness levels.

[Anonymous No.800377]

>>800351

4pcb.com (advanced circuits)

$33 2-sided, min 4 qty unless you're a student, min 1.

[HVAC/R guy No.800382]

::Starter Components Checklist::

[Resistors]

10K,1K - Often used as pull-up resistor for ICs with 5V supply
680 - Often used to limit current on an LED with a 5V supply

Potentiometer/Trimming resistor
10K, ¼ to ½ W - LM317 Trimming, other trimming purposes

[Capacitors]

Electrolytic Type (16V)
1000uF,680uF - For Output Filtering
10uF - IC Decoupling (High Current IC such as driver ICs), Voltage Regulator Filter

Ceramic Type
1uF - IC Decoupling (Low Current IC such as logic ICs)
0.001uF - Noise decoupling and for pin5 of LM555

Ceramic Disc 10%: 4.7pF, 10pF, 15pF, 33pF, 68pF, 100pF, 150pF, 330pF, 680pF, 1.5nF, 2.2nF,

3.3nF, 6.8nF (30): 22pF, 47pF, 220pF, 470pF, 1nF, 4.7nF

Multilayer Ceramic 10%: 15nF, 22nF, 33nF, 68nF, 0.22uF, 0.47uF (30): 10nF, 47nF, 0.10uF.
Aluminum Electrolytic 20%: 2.2uF, 10uF, 47uF, 470uF (30): 4.7uF

[Rectifying Diode]

1N4001 - for a maximum of 800mA
W02M - Bridge Type

[LED]

Red - Often used to indicate power or warning
Green - Used to signal good
7-segment common anode type - For displaying numbers

[Zener Diode]

5V 1/2W - reference, voltage clipper

[Transistors]

C636 - General purpose PNP transistor
C635 - General purpose NPN transistor

[IC]

74LS162 - Presettable Binary Counter
74LS163 - Presettable Decade Counter
74LS04 - NOT Gate can be configured as oscillator and digital filter or simply an inverter
74LS32 - OR Gate for summing
74LS08 - AND Gate enable / disable functions (bit masking)
74LS86 - XOR Gate selective inversion
74LS76 - Dual JK F/F
74LS47 - BCD to 7-segment Decoder (common anode)
LM555 - Multi purpose oscillator / one shot timer
LM324 - Comparator
LM358 - Op-Amp
LM431 - 2.5V reference
LM317 - Adjustable voltage regulator
LM7805 - +5V voltage regulator

[Anonymous No.800386]

>>800382
>680 - Often used to limit current on an LED with a 5V supply

No that's much too high. Common LEDs are normally 2v-3v 20mA - 30mA.

Take a 20mA 3v LED ("Super bright" green, blue, white).

With a 5v source you need the resistor to drop 2v. R = V / I = 2v / 20 mA = 100Ω

[HVAC/R guy No.800388]

::Some more resistors to collect::

¼W - 0, 1.0, 1.2, 1.5, 1.8, 2.2, 2.7, 3.3, 3.9, 4.7, 5.6, 6.8, 8.2, 10, 12, 15, 18, 22, 27, 33, 39, 47, 56, 68, 82, 100, 120, 150, 180, 220, 270, 330, 390, 470, 560, 680, 820, 1k, 1.2k, 1.5k, 1.8k, 2.2k, 2.7k, 3.3k, 3.9k, 4.7k, 5.6k, 6.8k, 8.2k, 10k, 12k, 15k, 18k, 22k, 27k, 33k, 39k, 47k, 56k, 68k, 82k, 100k, 120k, 150k, 180k, 220k, 270k, 330k, 390k, 470k, 560k, 680k, 820k, 1m, 1.2m, 1.5m, 1.8m, 2.2m, 2.7m, 3.3m, 3.9m, 4.7m, 5.6m, 6.8m, 8.2m, 10m

½W - 0, 1, 1.2, 1.5, 1.8, 2.2, 2.7, 3.3, 3.9, 4.7, 5.6, 6.8, 8.2 10, 12, 15, 18, 22, 27, 33, 39, 47, 56, 68, 82 100, 120, 150, 180, 220, 270, 330, 390, 470, 560, 680, 820 1k, 1.2k, 1.5k, 1.8k, 2.2k, 2.7k, 3.3k, 3.9k, 4.7k, 5.6k, 6.8k, 8.2k 10k, 12k, 15k, 18k, 22k, 27k, 33k, 39k, 47k, 56k, 68k, 82k 100k, 120k, 150k, 180k, 220k, 270k, 330k, 390k, 470k, 560k, 680k, 820k 1M, 1.2M, 1.5M, 1.8M, 2.2M, 2.7M, 3.3M, 3.9M, 4.7M, 5.6M, 6.8M, 8.2M, 10M

[Anonymous No.800408]

>>800388
in the time you find that, print it, cut it out, assemble it, tear it, make a new one, lose it...
you could just learn the colors.
hint: its basically the colors of the rainbow.

[Anonymous No.800421]

>>799197
Please respond.

[Anonymous No.800436]

>>800421

it's pretty obvious, if you go from 4 muxed LED digits to eight, then each one is only on half as much time, so it will be half as bright. but then, you decrease the series resistor and double the current, so you get back all your brightness. this can continue until you're pumping short bursts of 1 or 2 amps into the poor displays.

[Anonymous No.800437]

>>800421
I would if you made a minuscule amount of effort and posted the schematic instead of an unrelated picture and link to a pdf.

[Anonymous No.800443]

>>800374
>You can't use I2C for generic shift register interfaces
*generic* shift registers, but there are shift registers that use I2C, right? What's the problem with I2C? I wanted to cut down on the number of wires, at least down the road if I design a modular clock where the pairs of digits can move around. I really don't want 5 wires + 3 RGB wires between each segment. I'd rather have just 4 wires total.
Though I guess I should just focus on making a non-modular version first.

>An SPI port is just asynchronous hardware that will do this for you instead of an implementation in code (bit banging).
I don't know what that means. Doesn't SPI need the initial data from another IC just like I2C? {an IC like the ATmega 328 that's used in Arduino, but from what I gather the IC can be used without the Arduino board which seems to include stuff like USB interfaces or something.}

>WS2803
>It has eighteen outputs (enough for six RGB LEDs).
Is it possible to run 14 LEDs all using the same 3 RGB outputs from the 3 remaining outputs? {Man, I'm not even sure if 14 is enough. My original design does use 14, but I was contemplating adding a 15th dot to smooth a junction of bars, and/or removing 2. The 2 I was considering removing really need that 15th, I think; though luckily they could probably be paired for the on/off signal, unless that messes with voltage/current.}
This is where my poor electronic skills are showing, since I bet the capacity for numbers of LEDs per output has something to do with the volt/amp limits of the shift register. Ah great. I need to go back to the basics. At least I can find free online tutorials on that. just nothing about options for these chips.

[Anonymous No.800459]

>>800443
> there are shift registers that use I2C, right?
Yes, although they're normally called "IO expanders" Here's an example: http://www.ti.com/product/tca9534

Nobody really uses I2C to connect a microcontroller to a chain of shift registers. It's uses an addressable bus interface, so it can control multiple ICs through that.

What's the problem with I2C?
You're a beginner and it's more complicated than SPI. Don't make things more complicated than they need to be when you don't know what you're doing. You could find yourself in a situation where things won't work and you don't know if it's your I2C programming or the electronics.

> Doesn't SPI need the initial data from another IC
Yes, normally SPI is sent out of a microcontroller. Some microcontrollers will have an SPI port, but you can very easily generate SPI with code controlling any pair of IO pins.

> Is it possible to run 14 LEDs all using the same 3 RGB outputs from the 3 remaining outputs?
A WS2803 can not do this without additional hardware. Its outputs can only sink current. This means electrically they have two states: connected to ground, and not connected to anything (high impedance). This means the outputs must connect to the LED cathode and the LED anode must connect to VCC. This means you can not create a matrix because this requires control of LED Cathodes and Anodes.

WS2803s can be put into shift register chains. You can control dozens of them with one SPI port. Using a matrix will save you a few dollars, but it will make this project more complicated, which beginners need to avoid.

[Anonymous No.800467]

GE Transistor Manual 1964

http://www.introni.it/pdf/GE%20-%20Transistor%20Manual%201964.pdf

[Anonymous No.800478]

Threw together an astable multivibrator tonight. Seemed like it was time to learn how to do that.

[Anonymous No.800569]

>>800478
Now modify it to produce fast rising edges as well.

[Anonymous No.800822]

I work with a non-profit, I do tech'y stuff with kids.

I have ~$800 in my budget unspent for the year.

I have a CS background, and some experience with arduino and similar microcontrollers already. Are there some good general purpose breadboard kits that I could work on with kids? I dont know if anything requiring soldering is a good idea.

[Anonymous No.800825]

>>800822

plenty of choices; just pick one with a high rating.


http://www.amazon.com/s/ref=nb_sb_noss/186-7667780-2590354?url=search-alias%3Dtoys-and-games&field-keywords=in-One+Electronic+Project+Lab

[Anonymous No.800827]

>>800569
I always just put one of the outputs into another transistor. Works well enough. Here that is at 155 kHz. Rise time to 3.2V is roughly a microsecond. It doesn't get much longer if you use a higher voltage either. Not the cleanest thing ever, but it works for a lot of stuff.

Neat trick though.

Also
>owns an oscilloscope
>has never built an oscillator

:|

>>800822
I don't know about breadboard kits, but wireless power using an air core transformer could be really cool for kids. The circuits are relatively simple, and winding your own TX/RX coils is fairly easy (though the math can be a trick). Maybe if they're old/smart enough, it's an opportunity to teach them a little bit about inductance and electromagnetics.

>look mom, it lights up with no wires!
They'd go nuts.

[Anonymous No.800828]

Beginner here. What are some more recommended books I can add to my collection?

[Anonymous No.800838]

>>800825
I did some picking through those, they look cool, and I think one of those might be useful as a baseline for testing things but its pretty far from a complete project. I'm thinking of something long term that would keep them building, testing, adding new things.

So an Idea I had was to put together some arduinos maybe a rasppi and get some pushbuttons, hall effect sensors and set up a kind of mixing board ( we already have a 3d printer, so putting the whole thing together would be super cool ).

Drum machines, loops, the whole thing. There are just too many things you could do, because all you would need, is to create and modulate a sinewave, or loop samples from an SD card. it would be programmable but not difficult to work with.

[Anonymous No.800846]

I'd like to work with PIC chips and I've been thinking of the pickit 3 as it's the cheapest. However I've heard that it's very unreliable. Is it worth to buy and, if so, where should I buy it off from? Are the clones off ebay good to use like how the clone arduinos are?

[Anonymous No.800848]

>>800828
how many of them have you read?

[Anonymous No.800853]

>>800827

>owns an oscilloscope
>has never built an oscillator
>:|

Yeah I'm a mechanical engineering student, graduating this May. I just got into seriously dabbling with electronics and stuff, picked up that analog scope from some student up at Purdue last month for a mechatronics project I was working on for class. Learning more every day

[Anonymous No.800854]

>>800846
I've worked with the clone pickit2s a lot. It works as reliable as the real thing, but occasionally I'll kill it by accidentally connecting one of it's inputs to 12v (my projects often use 12v for motors and 5v for logic). This kills the clone. Real ones have protection against this.

[Anonymous No.800855]

>>800846

I worked with the PICKit3 in class. I never had serious problems with it, at worst it was unplug, plug back in. It's nice that you can have break points for debugging. IDK, anecdotal evidence but I see no reason why not to use it. Don't know about the clones however.

[Anonymous No.800859]

I'm rigging up 18 thermistors (9 air probes and 9 surface pads)

Do i really need 18 separate voltage dividers? There's got to be a way to optimize this

I'm new to this, pls be gentle

[Anonymous No.800864]

>>800859
That's really not so bad, dude.

[Anonymous No.800865]

>>800864
No, but it just seems like the type of thing where there would be a simpler method. It's fine if not.

[Anonymous No.800867]

>>800859
Making a voltage divider with your thermistor is just another resistor though. You could make an analog switch multiplexer and progromatically connect each of the thermistors to a single voltage divider resistor, but that's a lot more complicated than eighteen resistors.

[Anonymous No.800884]

>>799798
This is a fucking joke, let me see an analog maze solving robot

[Henry No.800885]

>>800884
>>799798
BUT FUCK IT

ROLLING

[Henry No.800889]

>>800885
shit already did that

It's a portable power supply with a big 12V battery inside, not exactly an Altoids container but it can charge anything from a battery to a laptop

[Anonymous No.800895]

>>799798
bored so gonna do a roll

[Anonymous No.800897]

>>800895
god no. reroll

[Swifts !WiNDUPyWhk No.800900]

>>800897
Hmm, that's an interesting one.

[Anonymous No.800904]

>>799412
Buying those kits is a good idea actually. I could have done it myself, but these kits are of a good price/quality which leaves more time for things I can not buy a kit for.

[Anonymous No.800941]

>>800443
You can chain an arbitrary number of shift registers, but a particular model of I2C IO expander usually only allows a maximum of four selectable slave addresses.

[Anonymous No.800967]

>>800859

if wiring up 18 resistors is too big a job for you, consider getting a resistor pack, like attached pic. this one has 8 pins, so it contains 7 resistors and one common pin.

[Anonymous No.801040]

Yo.

I'm trying to control a simple 3-pin computer fan using PWM on arduino and a N-MOSFET. I know it can be easily controlled using potentiometer without Arduino or PWM, but I would like to control it remotely.

It kinda works, but if I do analogWrite(0), which means always off, it picks up the pace and rotates faster. analogWrite(255) which is always on, slows down the fan, but doesn't stop.

What am I doing wrong? Why can't I stop the fan completely using PWM and why is analogWrite vice-versa?

[Anonymous No.801387]

>>801040
Bump

[Anonymous No.801624]

>>800884
What? No where does it say you have to make an analog maze solving robot. It'll be digital of course. The challenges clearly say to use your brain. If you need to use a microcontroller for the project then use it. Just don't be silly and use an arduino for a toy organ or something like that.

[Anonymous No.801630]

>electronics class
>professor starts talking about rectifiers
>"ripple"
>all I can think about is JoJo
why am I such a fag

[Anonymous No.801647]

>>799798
Roll

[Anonymous No.801671]

>>801040

Aren't 3pin fans DC fans?

Also, in the diagram you posted, you probably don't want to do whatever you're trying to do with the GND and the MOSFET like that. Common grounds are good things. Also, depending on what your power source is an how fancy you want to get, it might be worth your time to look into boost/buck converters.

[Anonymous No.801685]

What oscilloscope should I buy? Ideally below 400€, I don't want anything too fancy, it's mostly for hobby use since I'm an EE student.
I tried looking for used analog scopes in my area, but there are very few and ridiculously expensive, and my uni seems won't dump its dozens of old HP analog scopes anytime soon, so dumpster diving is not an option.

[Anonymous No.801707]

>>800388
Not sure why you'd want all those. Just the common ones would suffice.
1, 1.5, 2.2, 3.3, 4.7, 5.6, 6.8, 8.2

>>800408
Agreed
Black boys rape our young girls but violet gives willingly.
Get some now

[Anonymous No.801709]

>>801685

I have a USB oscilloscope that's functional enough for anything an undergrad EE will do.

[Anonymous No.801712]

Can I use a LM7812 regulator just for overvoltage protection? I know I can just use a zener diode, which is actually what the circuit i'm trying to build calls for, but I don't have any right now. Will the output be well below 12v if I hook it up in the regular way shown on the datasheet? If so, is there a different way to use the chip for what I want to do?

[Anonymous No.801715]

>>801709
This. Cheap ones will do, they're not inaccurate if they're well made but they're usually missing features that you can live without. And they run into problems if you starting measuring waves that are in the GHz range (depends on model, some start freaking out at lower freq), but I doubt you'll need to do anything like that.

[Anonymous No.801723]

>>801040
It would be nice if you add a 1k resistor on the mosfet gate (Arduino pwm pin.) and maybe add a pull down resistor between the mosfet gate (after the 1k resistor) and the gnd something like 10k would be nice..

[Anonymous No.801735]

>>801712
>Can I use a LM7812 regulator just for overvoltage protection?

nope. you want something that will eat any extra voltage by shorting it to ground. a regulator will not do that. I suppose it's possible to use a power transistor to do that, by biasing it such that an input voltage of 12V will apply 0.7V to the base thru a voltage divider, but it wont provide a sharp precise cut-off like a zener will.

[Anonymous No.801807]

Guys help. I am completely dense. Pic related is what I'm trying to make, but I don't have a 270 ohm resistor. what can i use instead? I'm using a 9V battery to light up 5 standard 5mm LEDs.

[Anonymous No.801814]

>>801807
What resistor values do you have?

[Anonymous No.801815]

>>801814
the ones mentioned below the black line

[Anonymous No.801817]

>>801807
>>801815
Not sure what E means, assuming ohms
Use the next HIGHER value, 330. (this LOWERS the current instead of increasing it, which could damage the LED's)

The brightness will be a bit less.
If that is a noticeable problem, you could use the two 560 in parallel for a resistance of 280 ohms. (560/2)

[Anonymous No.801819]

>>801817
E is Ohms.

last question, how do i wire it in parallel? As shown in the original picture?

Much thanks for the help.

[Anonymous No.801820]

>>801735
Okay, I've found a zener diode but it's not the value I need. Can I stack my one 12V zener diode with three normal diodes (forward biased) to achieve 15V?

[Anonymous No.801822]

>>801819
>do i wire it in parallel?
Wire circuit as shown, using 560 instead of 270
then take the second 560, and connect it across the first 560.

[Anonymous No.801833]

>>801822
Thank you very much dude. Much appreciated.

[Anonymous No.801834]

>>801807
You could probably add a third LED in the 270 Ohms resistor array, that would make the circuit more efficient, since the resistor is wasting 108 mW, and a third led would only waste 66mW (plus 60mW of another 150 ohm resistor) in conclusion, 20mW more, but more light.

If you can't do this, just add up resistors, like 100 Ohms and 68 ohms.

[Anonymous No.801838]

>>801834
I see. I'm building pic related. and it has a total of 5 yellow leds in operation. I wouldn't know where to direct the sixth one without having hot spots...

Hence working with 5 leds and the LED parallel calc gave me those values.

[Anonymous No.801840]

>>801834
also, any other way i could do this? Making the circuit more efficient with 5 LEDs then? Does that mean the 9V battery could power the circuit for longer?

[Anonymous No.801843]

>>801840
If you want efficiency, you'd need to minimize the voltage dropped in resistors. My suggestion is to ignore that and keep the circuit simple.

If you have no use for six leds, you could always paint the extra led black or use heat shrink tubing or something. That wouldn't be more efficient, but all the 5 visible led would track each other's brightness better as your battery depletes. Dunno if that actually matters.

[Anonymous No.801848]

>>801843
Ok. I'll just keep it as is. It's for display purposes only so shouldn't matter much. Thanks for your help.

[Anonymous No.801850]

>>801838
>>801840
You're good. Use the circuit you have

Efficiency is just comparing the actual useful power used (light) to that wasted(heat)
Adding another LED that you're not using will do nothing for battery drain, just less power "lost" by resistors. Since a 6th light would also be a "loss", it makes no difference.

The only real difference is that when the battery drains, the branch with 3 LED's will go out first. So, I'd put the 2 LED's with the 2 resistors front (or sides) for symmetry which you were probably going to do anyway.

[Anonymous No.801852]

>>801850
Cool that clears up a lot. Thanks for helping!

[Anonymous No.801853]

>>801852
Btw, if you want the battery to last longer, the easiest way would be to use more efficient LEDs and drop the current used to feed them. The modern LEDs produce much more light at the same current than the ancient ones.

[Anonymous No.801856]

>>801853
I'll be over the moon if I could just get this to work. So for now, till i'm better at this, i'll go with what i have.

I could use your help in another project. You game?

[Anonymous No.801859]

>>801856
I'm a regular here, but me being here or not shouldn't really matter. If not me, someone else will answer your questions.

[Anonymous No.801862]

>>801859
yeah i tried that once. I got torn apart. Mostly my fault cause i had no idea what the hell i was talking about but whatever.

[Anonymous No.801866]

>>801862
Not him but that's the exception here.
Your questions in this thread might get you shit due to /g/'s elitism bleeding over. I'd help you If you created a thread with basic questions.
Anons here bitch if they feel you haven't tried and are just saying "solve this problem for me". I helped anyway because it was obvious that this was for another project and the electronics was secondary.
Maybe sharing the larger goal and make it interesting would get you a better reception. Give something in return.

For example: I felt better about helping when you posted the pic. >>801838

[Anonymous No.801887]

>>801709
Dunno if you're still around, but this. I think you get a discount if you're a student.

http://www.digilentinc.com/Products/Detail.cfm?NavPath=2,842,1018&Prod=ANALOG-DISCOVERY&CFID=9093561&CFTOKEN=f0a045f058f8daf5-97D9C88A-5056-0201-028101B244EBA154

[Anonymous No.801894]

>>801866
actually the problem is i'm just getting into electronics. That too basic lighting and shit and everything is confusing as fuck. I got Forest Mims book and one from Make magazine but it's slow going. The Led Calc was a revelation by giving me a shortcut. I downloaded Fritzing, and Eagle and i'm trying to work through them. Got a starter electronics kit, ordered a bunch or resistors of ebay but i'm still as lost as ever. So I agree with all your points and thanks for all the help

[Anonymous No.801898]

>>801894
Seen this recommended before and it looks very similar to the material and order I learned in tech school.
allaboutcircuits.com
Start with Direct Current

[Anonymous No.801916]

any of guys worked with PLC before?
im helping my dad out with his company and currently the customer is a carwash who wants to centralize their error handling into one device instead of sensors that just beeps when shit happens
the PLC was easy enough to program but the customer wants to be able to remote control shit using the internet if possible (turn the water or electricity into a specific booth with the press of a button)
im pretty sure this is impossible using only PLC but is there some other way to do it?

they also want to log down errors into a server/file if possible

[Swifts !WiNDUPyWhk No.801919]

>>801916
What kind of a PLC are you using?
I haven't done any real work, but we're learning how to use them in school.
I do however know that what the customer wants is possible with the PLC we use, which is Siemens Simatic.

And I'd honestly be surprised if any sort of industrial PLC shipped without some sort of networking support on it.

[Anonymous No.801926]

>>801919
granted, im using a pretty old mitsubish FX1S for practice right now so that might be the issue
im kinda scared to order a new one without knowing exactly what it's capable of though, its a lot of money after all

thanks for your help anyways, ill look into some of the modern models and what they can do, preferably without using a HMI

[Anonymous No.801940]

>>801916
From personal experience I can say that most ABB PLC's have integrated TCP/IP stacks so you can just Cat5 them up to your router or switch or whatever. Plus, they allow you to set up a log (be it errors or parameter trends etc.) in the built in flash or external SD card which will be formatted on the fly (FAT16 if it matters to you).

I'm pretty sure most modern PLC manufacturers have support for those things too. If you can't get one, connect the existing one to a PC and run the needed software there.

[Anonymous No.801958]

>>801685
rigol ds1054z

Can be hacked to get full 100mhz and misc options

[Anonymous No.801966]

>>801820
>Can I stack my one 12V zener diode with three normal diodes (forward biased) to achieve 15V?

you'll get a 13.8V zener and the zener action will be a little mushy (not as sharp) as with just a zener, but yeah.....

another way to do this is with a power transistor. see pic. say you wanted to limit a voltage to 15V but only had a 12V zener. you'd arrange the voltage divider to provide 12.6V at the junction, so the transistor would turn on and short out any voltage above 15V on the collector.

[Anonymous No.802050]

>>801966
Thanks. I assume this is going to work better than stacking diodes? Can I put several transistors in parallel to increase the amount of current it can shunt to ground without blowing up?

[Anonymous No.802056]

>>802050
Depends. The primary advantage is that you can use a power transistor instead of a power zener. High power transistors are cheaper and easier to get.
Paralleling bipolar transistors is bit iffy and in practice you'd need emitter ballast resistors to prevent one transistor from hogging all the current. Use one big transistor instead. Consider using a darlington transistor, if the required current is high.

[Anonymous No.802063]

>>802056
In the case of the darlington I'd need the voltage divider set at 13.2v, right?

[Anonymous No.802065]

>>801958
Neat, I knew about the rigol brand, but I had not discovered that model yet, after looking at Dave's reviews I'm impressed, specially for something under 400€, 4 channels! The display feels very analog like, I hated the last versions in which the waveform had the same intensity for all the area. The hack capability is also a huge advantage.

Thanks man!

[Anonymous No.802081]

>>801966
I think I'm actually going to go with the stacked diodes. This is going into a car and I don't want it to be drawing any current when it's not running. The leakage current of a diode is way less than anything I can get out of a voltage divider without using megaohm resistors

[Anonymous No.802224]

>>800828

Start by removing 95% of those and reading the rest.

[Anonymous No.802288]

>>802063
Yes.

>>802081
Sure, makes sense, but if you worry about that, check how much current your diode-zener combo draws when it's "not conducting".
Also, the self discharge rate of the car batteries is high enough that you can draw some milliamps without any noticeable effect.

[Anonymous No.802349]

>>801887
I think I'll rather go with the rigol ds1054z. There's a lot of documentation about it since it's quite popular, and it's also quite compact. Usb oscilloscopes aren't really my thing.
Also:
>the customer must be a student at a US academic institution
>yuropoor
Thanks anyway!

[Anonymous No.802394]

Multimeter suggestions around 25 bucks?

[Anonymous No.802476]

bump

[Anonymous No.802482]

What is the minimum voltage to an op amp comparate when it is configured as a comparator? I was playing with a NE5532 and when I was getting something like 300mv from my voltage divider, my op amp stoped comparating the inputs properly. The schematic I used is the one on the picture.

[Anonymous No.802499]

>>802482
probably around 300mV for most op-amps as you've discovered. The datasheet should have the exact value.

[Anonymous No.802513]

>>802482
Depends on opamp or comparator. See the datasheet. It's listed under "common mode voltage range", CMV, or something similar.

[Anonymous No.802750]

So I'm dicking around with transistors, and I'm getting this on my scope. It happens maybe 2 to 5 times every 10 seconds. It's too short and irregular for the scope to trigger. Will post circuit in a second.

[Anonymous No.802751]

>>802750
So what's going on here? Is it appliances switching on and off? Lightning? Aliens?

[Anonymous No.802756]

>>802751
It appears you've made a high sensitivity electrometer. It's reacting to the movement of charges in your table; possibly caused by yourself moving about.

[Anonymous No.802758]

>>802751

Maybe charge builds up in the 90nF cap and then opens the base, discharges and starts again. The base can draw large currents without a resistor. Judging where you placed the probe, I would expect to see a line at 4volts, and then a drop in voltage near to 0 when that thing happens. I don't know why the line is at 0 (say no offset) and why all that noise appears. Maybe it has to do with the steel acting as antenna. A video of the signal would be appreciated, even if it's just to contemplate the beauty of CRT oscilloscopes.

[Anonymous No.802761]

>>802756
Maybe adjusting the resistor you could make the pulses stop and trigger the transistor by touching the table, since the transistor could be at edge.

[Anonymous No.802763]

>>802756
I do have my little LM317 box and my scope sitting on the same table. The scope chassis is grounded and sitting on rubber feet though. What could I do to trigger a response? I tried switching lights on and off, and it didn't do anything.

I thought that it might be responding to something I was doing, like shuffling around on the carpet. So I sat in a chair across the room and kept still, and it didn't seem to change anything. Either way it's really neat to watch the scope trace dance around. Makes me feel like I discovered something.

I was originally trying to pick up lightning strikes, so I guess I got kind of close.

[Anonymous No.802764]

>>802758
I'm trying to make a webm but apparently I'm dumb. Also the scope is AC coupled. If I DC couple it, it sits above ground, which makes perfect sense.

>>802761
I had to do quite a bit of fiddling with the resistor to get anything to come through. The window is very small.
If I take the cap out, there's no signal at all

[Anonymous No.802785]

>>802764
>hopefully this posts

[Anonymous No.802852]

>>802785

reminds me somewhat of a gravity-wave detector I built a few years ago. it used the surface plates of a capacitor to (theoretically) detect theses waves. so, maybe the big metal plate is capturing something unworldly.

http://www.keelynet.com/gravity/grav3.htm

if I recall, if you swung a large metal object around in a circle near it, the circuit (a diff one from the one in the link above) would react to it.

[anus presley No.802853]

>>802785

reminds me somewhat of a gravity-wave detector I built a few years ago. it used the surface plates of a capacitor to (theoretically) detect theses waves. so, maybe the big metal plate is capturing something unworldly.

http://www.keelynet.com/gravity/grav3.htm (this is a diff circuit than the one I built)

if I recall, if I swung a large metal object around in a circle near it, the circuit would react to it. spooky!

[Anonymous No.802886]

Hello, /ohm/.

Considering the pic related I want to know if there is any circuit I can do to conect this 12 volt relay on a 24volts suppply. I cannot use a voltage regulator and a resistor will get too hot.
Is there any circuit I can do using transistor or zener diodes?
Thanks in advance.

[Anonymous No.802904]

>>802886
Get a 24V relay instead. Or get a big enough resistor. Or provide adequate cooling for whatever thing you're using.

Everything except a step down switching regulator will waste the same amount of power. How hot things will get depends on their size and cooling.

[Anonymous No.802950]

>>802886
Two relays in series?

[Anonymous No.802974]

>>802886
does the relay run on 12v or handle 12v?

because if it runs off 12v you can simply use a 12v regulator

even simpler
power relay with series resitor of value
(24v-12v) / (current draw of relay in amps)

of course you should go up one standard value from whatever that formula yields

make sure you select one with the correct wattage rating of course too.

[Anonymous No.803013]

>>802974
>>802904
>>802950

The relay must have a 12V coil, I cannot use a regulator and I want to use a resistor as last option.
Isnt there any circuit I can make using a transistor or maybe a zenner? I cannot find one on google, maybe my search keys are not the correct ones.

[Anonymous No.803026]

>>803013
You can use a 12V zener in series with the coil, but it's going to produce just as much heat a resistor would and zeners are more expensive and more difficult to cool. Same with any simple transistor circuit.
You could reduce the power dissipation by paralleling the resistor with a capacitor and dimensioning the resistor so that it provides only enough current to keep the relay closed. That would be a shitty kludge, though.

>The relay must have a 12V coil
Why?
>I cannot use a regulator
Why? A switching regulator would to what you're asking without producing lots of heat.

[Anonymous No.803034]

>>803026
Because it is a 12V relay with contacts for 40 amps. I did not found a relay for 24V with the same contacts specifications in my city and I don't have time to order it on the internet. A regulator surely would resolve the problem but I was wondering if there was another way to solve the problem.

[Anonymous No.803050]

>>803034

any solution will involve putting something in series with the relay to suck 12V out of the 24. pretty much every reasonable thing has been suggested already except maybe a lamp from a car of the appropriate wattage, or 11 diodes in series.

(out of curiosity, is the 12V relay from Adisson?)

[Anonymous No.803051]

>>803034

any solution will involve putting something in series with the relay to suck 12V out of the 24. pretty much every reasonable thing has been suggested already except maybe a lamp from a car of the appropriate wattage, or 17 diodes in series.

(out of curiosity, is the 12V relay from Adisson?)

[Anonymous No.803178]

>>800408
Bad Beer Rots Our Young Guts But Vodka Goes Willingly

[Anonymous No.803249]

>>803051
>17 diodes
I lol'd.

[Anonymous No.803294]

I've been thinking of buying a power supply to power this project with a limit of 1.5A and I've found this to be a good solution:
http://www.superpowersupply.com/index.php/super-power-supply-ac-dc-adapter-charger-cord-9v-0-5a-1a-5-5mm-x-2-1mm-2-5mm-wall-barrel-plug.html
The problem is that it says it doesn't give a expected current. Does current change depending on the load receiving it? And, if so, will this mean that the load will never receive too much or too little current and won't damage the parts?

[Anonymous No.803302]

> Does current change depending on the load receiving it?

yeah, dude, dont they teach physics in high school anymore? the formula is called ohms law: current = voltage / resistance

image a lightbulb in a socket. the socket can supply as much as 15A, but it doesnt mean any bulb you screw in will take 15A. it depends on the resistance, as per the formula above. a 60watt bulb, for example, will only take 0.5A.

[Anonymous No.803308]

>>803294
I don't get why it lists 2 different currents (0.5A &1A).
Current rating on power supplies is a maximum rating, not to be exceeded.
If your project requires 9Volts and 1.5Amps, the supply is inadequate if it's rated 1A

>Does current change depending on the load receiving it?
For a fixed voltage, yes
This means that the supply wont deliver the required current and/or may be destroyed.

> project with a limit of 1.5A
You want a supply that meets or exceeds the current(amps) requirement.
9V 1A = insufficient
9V 10A = overkill but fine.

[Anonymous No.803311]

>>803034
The simplest (but not very efficient in size and cost) way would be to use two of the 12V relay coils in series.

[Anonymous No.803346]

Where do you guys like to buy your electronic components?

Right now I'm looking for DC motors.

[Anonymous No.803354]

>>803034
You could build some kind of PWM thingy for driving it, but PWM + relay isn't really that much different from a switching regulator.
It does not need to be particularly complex. For example, a 555 astable + mosfet or even two-transistor astable + darlington transistor would do the job.

[Anonymous No.803362]

First off sorry for barging in like a retard with no idea what I'm doing.

Now, I want to emulate a PC fan tacho, I've found this little circuit (pic related) from this site http://www.techidiots.net/notes/fake-fan-sensor

My question to you guys is, can I put the fan back in that circuit, and is the Texas Instruments SE555P suitable to use as the 555 timer instead of the LM555N I see there?

Ordering the parts from RS, minimum order for LM555N is a tube of 50, but you can get SE555P in a pack of 5.

[Anonymous No.803363]

>>803362
too retarded, forgot the pic.

[Anonymous No.803365]

>>803362
>can I put the fan back in that circuit
In what sense? You can use the same connector for powering the fan, if that's what you meant.
>is the Texas Instruments SE555P suitable
Yes.

[Anonymous No.803367]

>>803365
Thanks re. the SE555P.

This is what I mean by adding the fan back in, excuse the crude drawing. I believe the correct way to describe this is as putting the fan in parallel?

[Anonymous No.803378]

>>803367
You can out it back like this, no problem at all.

[Anonymous No.803380]

>>803367
You can put it back like this, no problem at all.

[Anonymous No.803405]

>>801822
>>801807
>>801838
Guys this is me again. One last question. I got myself a switch like pic related. It has three prongs. How do i integrate it into the circuit? And which wire do i connect to which prong?

[Anonymous No.803409]

>>803405
The middle terminal connects to one of the other two depending on the switch position. Usually the one opposite of switch direction.
So, any two adjacent terminals.

Put it between the battery+ and circuit

[Anonymous No.803410]

>>803405
The middle pin is the common (com in pic related) and it is connected to either of the remaining pins, depending on the lever position.
Cut the wire going to the battery (no matter which side), connect it to the switch middle pin and the remaining wire to either of the side contacts.

[Anonymous No.803412]

>>803409
>>803410
Awesome. Thanks guys!

[Anonymous No.803425]

>>803410
>>803412
Good rule of thumb is to always switch the "hot" side of a circuit power supply, thus ensuring it is completely de-energised and disconnected from supply voltages. In a DC system positive is considered the hot (power bearing) line, so the switch should be inline with Battery+. Not so important with battery powered projects but much more so in AC land, may as well get in the habit now.

[Anonymous No.803473]

what are some fun beginner projects i can do with an FPGA demo board?
no blinky lights please

[Anonymous No.803492]

>>803473
Oblig: stick it up your butt.

Build a high-speed stock trading platform that makes decisions in 100 ns via neural networks and HW-accelerated TCP/IP stack. Blow up the economy. Get sent to jail. Get ??? stuck up your butt.

[Anonymous No.803519]

>>803492
pls

[Anonymous No.803528]

Does anyone have experience in trying to manufacture PCBs with a CNC engraver? I'm trying to turn my design into G-code, either directly or by exporting Gerber files first, but it's a fucking nightmare. Eagle's pcb-gcode script exports only a partial toolpath for whatever assbackwards reason, FlatCAM looks good but the actual code is horrible with every movement being a sub-1mm step, and everything else I've found is even more dire. Any hints?

[Anonymous No.803537]

>>803528
There's (at least) one guy here who uses an engraver/mill to make PCBs, but afaik he more or less draws his layouts as tool paths.

[Anonymous No.803546]

I'd just like to make sure but does wire on components have lead? I touch them all the time with breadboard and the thought just came to me. Also is it bad to just touch solder or is it only dangerous once I use it with an iron?

[Anonymous No.803548]

>>803546

most solder is lead-less now. even if you find solder with lead, you're pretty much fine as long as you don't eat it or something.

[Anonymous No.803553]

>>803537
Sounds like you're talking about me.

I use Eagle to design PCBs like a normal person sometimes, but I can get things a lot more optimized by drawing the tool paths directly. I do this by drawing on the dimension layer then exporting a cut instead of milling. The main issue is that the gcode milling just draws an outline of everything on top layer. This means if there's parallel lines running next to each other it generates a toolpath that cuts the copper between them twice. This makes everything take twice as long and can make the remaining copper traces too thin.

When generating gcode the normal way or my special way I use this ulp instead of the standard pcb-gcode: http://pastebin.com/bsHMAac0.. It's a lot simpler and easier to use.

I also use Gopt available here: http://www.cnczone.com/forums/pcb-milling/88972-gsuite-free-gcode-tools.html

The gcode generator I linked and the standard one both generate toolpaths based on the order you drew the lines which means the machine will have to jump all over the board in an inefficient pattern.

[Anonymous No.803559]

>>799798
Sure, why not, ROLLIN

[Anonymous No.803564]

>>803548
So touching it and not washing hands is fine?

[Anonymous No.803568]

>>803546
All soldering does is burn off the flux core. It's 40/60 lead before and after. You don't absorb lead by touching a solid piece of it though; just wash your hands between soldering and eating.

The lead free ROHS exists mostly for environmental reasons. The world generates 50-million tonnes of E-waste a year and a lot of it isn't disposed of responsibly. Your hobby projects aren't generating much E-waste and you have to be pretty stupid to get lead poisoning.

[Anonymous No.803641]

>>799798

roll

[Anonymous No.803642]

>>799798
rerolling, already made one in school

[Anonymous No.803649]

I want to control this motor
https://www.pololu.com/product/2275/
.

Ideally I would like to find a motor controller that can output some amperage above the stall current (6A), but I can't find any controllers whose circuits are physically small enough (~0.8 inch x 1 inch max footprint).

Could I get away with using one of the smaller boards like this one:
https://www.pololu.com/product/1212

I expect to be requiring torque about 1/2 to 3/4 that of the stall torque if that helps.

[Anonymous No.803658]

>>803649
Speaking of motor controllers, if it has an H-bridge, direction can be controlled, right?

[Anonymous No.803661]

>>803658
yes.

[Anonymous No.803662]

>>803658
yes, that's the purpose of an H-bridge

[Anonymous No.803769]

>>799798
wolling

[Anonymous No.803922]

>>799798
roll

[Anonymous No.803929]

If you want to do anything with handmade digital computers, I can't recommend the textbook "Digital Computer Electronics" enough. It's really great

[Anonymous No.803931]

If I have a standard beeper-thing (pic related) without an oscilator, what is the absolute easiest way get it to make a sound? I don't care about what kind as long as it's audible. There has to be some tricks besides making a complex sine oscillator, right?

[Anonymous No.803934]

>>803931
0. Drop it on the floor.
1. Connect it to PC/cell phone/whatever's audio output

[Anonymous No.803936]

>>803931
555 astable circuit.

[Anonymous No.803966]

>>803931
555 IC

[Anonymous No.804038]

>>803931
You can try this, it's not like it's rocket science, change the capacitor value to change the frequency. Lower voltages can be used.

[Anonymous No.804044]

>>803931
Here's a simple oscillator design for 1khz. Getting it to actually work is a little tricky. Use a 1k potentiometer in series with the 22k and adjust to get it to oscillate.

[Anonymous No.804459]

>>799233
This might not be the right thread to ask this, but can IR LEDs bleed through semi-opaque items like certain keycaps? I've been dabbling with the Leap Motion in in conjunction with my DK2 of the Oculus' Rift and have been contemplating assembling a 60% mech keyboard whose keycaps are backlit with IR LEDs for pic related use.

[Anonymous No.804467]

>>804459
things can be transparent at different wavelengths
like how you can't see through walls but you get wifi signal.
you just need the right material

dont ask me if it exists though.

i know anti-static bags work exactly the opposite way though

[Anonymous No.804483]

>>804459
even if you can't find caps that are ir-transparent (which i'm thinking abs and pbt are not) you can edgelight the key groups with leds inside the casing, maybe that'll be good enough

failing that, i bet the visible-transparent caps you see on cherry rgb-backlit boards are ir-transparent too, but good luck finding a set

[Anonymous No.804503]

So I am going to drive a brushed DC motor according to signals from an ATtiny. I consider building and H-Bridge for this. But also, I am looking into ICs but can't find a chip that can handle the amperages. The motor will be driven with ~6V and should draw about 2A max. All the chips I found can't handle more than .5A, except the IXDN404 which supposedly handles up to 4A. However, I read in some guy's blog, that it gets uncomfortably hot at .4A already. So, my question is: Are there immensely popular motordriver ICs out there that I missed and what are their capabilities?

>tl:dr: Motor drivers

Thanks for giving a crap!

[Anonymous No.804521]

>>804503
Use 555s.

[Anonymous No.804534]

>>804503
just use the ATtiny to directly drive some logic level MOSFETs, as you've already found there are no decent drivers except for low current applications.

alternatively if you are dead set on using a controller because you can't program or something you can just use the controller to drive said MOSFETs to make up for it, you will have to modify some component values of course if the driver has current limiting and feed back sensing.

[Anonymous No.804825]

Might as well post this here instead of making a new thread.

I'm new to electronics and I've got a project I'm currently working on that involves sensors. I'm using a diode/transistor setup for this but I'm having trouble calculating appropriate resistor values to use.

According to the datasheet (which I'll post as a reply to this) the maximum ratings for the diode are 2V and 40mA. If I was to use Ohms law to figure out the resistance required to get these values across the diode would I be able to just use R = 2 / 0.04? I'm not sure which value to use for V when I have the input voltage as 5V and the desired voltage as 2V.

The current resistor values in pic related are just placeholders.

[Anonymous No.804827]

>>804825
Datasheet as mentioned

[Anonymous No.804838]

>>804825
Find the voltage across the resistor then use ohms law R = V/I

R = (V_supply - V_diode) / I_diode
R = (5V-2V)/40mA
R = 75 ohms minimum

[Anonymous No.804851]

>>804838
Thanks anon

[Anonymous No.804884]

>>804534
alright, I was thinking about that anyways. So I got an AtTiny 44-20PU (I need more than 8 Pins) and I am going to get the PWM directly from the chip (programming, WOO!). I found two transistors, the STP 16NF06 (NMOS) and the IRF 9520 (PMOS). They have exeptionally low switching times (around 20ns) and apparently can handle upt to 65V/5A or something. Would that suffice? I really don't know what I have to look at in transistors apart from a high switching frequency, since I would like the PWM to have a frequency of 20kHz or more so it doesnt ring like and ass.

[Anonymous No.804906]

>>799798
roll

[Anonymous No.804919]

Hey there /ohm/, first time poster here...

I have an idea for a project, I want to start looking into it by myself, but I wanted to ask here before in case it's straight up retarded so I don't loose time looking up info in something that is straight up retarded.

I want to build a portable motion sensor, strip some Chinese PIR's to salvage all the motion detection part, but instead of it emitting and acoustic signal rewire that to transmit a signal onto a certain frequency, UHF or VHF, buying whatever components I need to do so, range wouldn't have to be incredible, so any other transmission method would work as well... I was looking at some old WI-FI antennas and came up with the idea, so bare with me if this is straight up retarded.

Thanks in advance

[Anonymous No.804924]

I don't think this deserves a separate thread and it has to do with electronics so here goes:

Potentiometers: what's the correct way of installing them on a panel?

I've seen pics of nut/washer/panel/washer/nut and just panel/washer/nut. The first seems like overkill, but I haven't mounted them on panels before so I want to make sure...

If it matters, there are 8 potentiometers and they are also soldered onto a PCB inside the box.

[Anonymous No.804929]

>>804924
Honestly, for the majority of the time, gluing them on works fine for hobby projects, although for professional projects, you will need to do what you mentioned.

[Anonymous No.804948]

>>804924
you usually put a washer on where a screw or nut is turning against something that you don't want to mark or scratch or when you need to spread the load out if your material is too thin.

the stress from a potentiometer on a panel doesn't need a washer for spreading loads but if you have pot > panel > nut the nut will scratch the panel so you want pot > panel > washer > nut so that the nut is rubbing the washer not scoring your panel.

if the panel is too thin you might not want the pot poking out too far so you can add washers as packers, pot > washer (as many as you want) > panel > washer > nut.

really you aren't building a bridge or a building, nobody will die because you used too many or too few washers.

and then you have spring washers, washers with indexing pins, nylon locknuts... fixings hardware is fun

[Anonymous No.804962]

Can't figure out why Saleae logic isn't identifying these i2s signals. Settings or timing issue?

[Anonymous No.804965]

>>804467
>>804483
>even if you can't find caps that are ir-transparent (which i'm thinking abs and pbt are not) you can edgelight the key groups with leds inside the casing, maybe that'll be good enough
Someone pointed me to translucent keycaps; http://www.maxkeyboard.com/clear-translucent-cherry-mx-blank-keycap-set-for-esc-w-a-s-d-or-e-s-d-f-and-arrow-keys.html
But ye when push comes to shove I could always do the latter option you mentioned.
>bet the visible-transparent caps you see on cherry rgb-backlit boards are ir-transparent too, but good luck finding a set
I recently found doubleshot backlit keycaps are a thing; http://mechanicalkeyboards.com/shop/index.php?l=product_detail&p=1003
So I guess they're not very hard ti come by.

[Anonymous No.805003]

Hello Ni/diy/erians,

Imma a little rusty how do I express 4.7 micro farads as a whole number is it like this 0.000047?

[Anonymous No.805007]

>>805003

you joking nigga?
the first 3 digits are after the dot are thousandths: so 000 thousandths
the next 3 digits are the millionths: so 047 millionths

so your number is 47 millionths or 47 micro.

[Anonymous No.805011]

>>805003
http://lmgtfy.com/?q=farads+in+4.7+micro+farads+

[Anonymous No.805591]

FINALLY FUCKING GOT IT
https://www.youtube.com/watch?v=SyiIiXdw8_E

earlier thread:
>>797344

runs off arduino

looks sloppy, but it does exactly what we want. we'll clean up the wiring.

at least we're gonna fucking graduate.

[Anonymous No.805594]

>>805591
(The idea is that you flip the "enable" switch to allow the motor to move, then the second switch orders it to "chase" a given direction, decelerating when it trips a hall-effect sensor near either end.)

[Anonymous No.805639]

I come to you guys to beg like a little bitch. I have built the fucking sine wave circuit five times now and each time it spits out a square wave and flashes an LED instead of fading it in and out. At one point I got it to fade up, but not down. I don't want to use the Fading LED circuits right now because I wanna do it with as few parts as possible. Every goddamn time I try the transistor method, it just flashes instead of fading.

Can any blessed anon help a poor soul like me? I just want to build a sine or triangle generator with a 555 timer and a few resistors and capacitors.

I have read shit all over the place and I keep finding people who say "Oh just order this obscure IC and save yourself the trouble." Fuck that. I just wanna shave a square wave.

Thank you guys in advance.

Pic related.

>Yes I mad.

[Anonymous No.805646]

>>805639
You're probably not setting the RC time constant correctly, but it's probably going to give you a shit result no matter what you do. A square wave is made out of fundamental and all odd harmonics, so you need to filter out everything except the fundamental. A RC filter is first order, -20 dB/decade, -6 dB/octave. The 3rd harmonic is going to be reduced by eh 10 dB let's say at best.

[Anonymous No.805651]

>>805639
>it spits out a square wave and flashes an LED instead of fading it in and out.

Do you know it's a square wave by looking at it on a scope?
LEDs don't dim with voltage the way incandescent bulbs do. They have a very narrow operating voltage. A quarter volt below the rating and they don't light up at all. A sin wave won't produce a smooth dimming effect, it will go from off to on very quickly when the signal voltage gets near the rating. This is why LED dimming is normally done with PWM.

> At one point I got it to fade up, but not down.
Because the capacitor you're using in the bottom right is discharging through the LED. These circuits were not meant to drive any kind of load directly.

[Anonymous No.805654]

>>805639
An LED's intensity is dependent on current, not voltage. Therefore you cannot make an LED fade by adjusting the voltage. You can either adjust the current (e.g. use your sine wave as input into a current regulator), or use a PWM.

[Anonymous No.805655]

>>805639
You should be able to do PWM with an adjustable duty cycle using a 555 and a variable capacitor. (or possibly two 555s, one in monostable mode) Try that.

[Anonymous No.805659]

>>805651
>>805655

Thank you so much for your replies. What would be the best way to do Pulsed Width Modulation that fades an LED?

[Anonymous No.805660]

>>805651

This makes so much sense as to why my LEDs would flash instead of dim. Thank you for your reply.

[Anonymous No.805719]

>>805659
>What would be the best way to do Pulsed
Width Modulation that fades an LED?

I have a better idea. you use a 4017 counter like in the pic, but instead of lighting 10 LEDs, you just light one using 10 diff resistors. the first 5 resistors are in decreasing order to make the LED get brighter, and the last 5 increase the resistance to make it fade. you might leave the first and last outputs unconnected so it's completely off for 2 counts. the 555 determines the rate of the fade.

[Anonymous No.805854]

What is the power rating of some standard earphones?
I'm bulding a sensory amplifier with 2x 741 OP amps and a 3904 for the final amplification and driving the earphones. I don't want to fry my sennheissers, how much current could they take in quiescent mode? Which would be the max current so I can limit it putting in series with a resistor?
Also, does it matter if the OP amp has just the positive input enabled while the negative is connected to ground when amplifying signals in non inverting mode? Should I bias the microphone so the signal can go up and down in the positive range?

[Anonymous No.805868]

>>805854
They have an impedance of 16 Ohms btw, but the power or voltage/current range is not specified.

[Anonymous No.805923]

How would I go about creating a continuous adder?

For instance, how would I take the circuit in pic related and loop the output back through the adder?

[Anonymous No.805931]

>>805003
0.000 004 700
milli, micro, nano, pico

[Anonymous No.805932]

>>805923
Wat?
Make another exactly the same as this but connect cout to cin of the other one.
Repeat ad nauseum

[Anonymous No.805950]

>>805932
No like loop the output back into the same adders

[Anonymous No.805965]

>>804919
just use an RC transceiver/receiver. Google that shit up yo

[Anonymous No.805966]

>>805923
>>805932
Whatchur lookin for there is a counter.

[Anonymous No.806027]

>>801630
because your class has been funded by the best research organization in america

[Anonymous No.806046]

I've been seeing if my nixie tubes work and I've stumbled on a few problems. One is that some digits dont light up properly, even with a 125V input, though sometimes they do light up fully in a certain position. How can I make them fully light up? Another one is that this strange blue mass appears in certain positions. What is this?

[Anonymous No.806056]

>>806046
Actually a nixie tube requires about 170V. Try using that and the digits should fully light up. I've got no idea what the blue lights are, but maybe it's related to cathode poisoning? Also that is one ghetto setup you've got there

[Anonymous No.806058]

>>806046
Isn't that fairly endemic to nixie tubes?

[Anonymous No.806061]

>>806056
Perhaps. I'm just getting started on this and don't want to fuck around too much though, so I'll try seeing what happens when my second HVPS arrives.
>>806058
I've tested about 8 of these and only 2 had this.

[Anonymous No.806080]

>strange blue mass appears in certain positions
that means you're applying ac to the tube. they need dc to light up correctly without producing that blue arc.

[Anonymous No.806083]

>>805950
Connect the sum outputs to one of the input ports. Put 4 flip-flops in between, so that it increments only when you give it a clock pulse.

[Anonymous No.806132]

>>806046
That blue mass might be air, check if your nixie has a crack, it might also be mercury, some add mercury to make them brighter.
>>806056
I agree with him, you could just rise the voltage, even if that implies having to limit the current.
>>806080
If it was AC wouldn't the grid light up too?

[Anonymous No.806184]

>>806132
Turns out a higher voltage fixed it. However I fucked up and connected the lead to a comma cathode and it zapped and turned blue instead of the usual neon orange. Nothing bad seemed to have happened but is it really alright?

[Anonymous No.806195]

>>806184
I just tried doing this again. Some smoke appeared from near the battery but, after doing it again a few times later, everything is still fine? Will it be okay?

[Anonymous No.806235]

>>806195
Are you even limiting the current? It doesn't sounds right to me to put that many volts (to what's basically a neon lightbulb) without a resistor. Maybe the current flowing through the neon is too big and it causes even more current to be drawn from the battery, stressing the step-up circuit.
This said, I don't know much about nixies, I just think it's the most obvious problem.

>>805854
I found out how to bias my lm741 OP amp so the whole waveform shape coming from the mic could be conserved and amplified. However I need to know how not to fuck up my earphones. Googling a bit has shown me the output from a standard jack can peak 3 volts ( from -3 to +3).

Now, I don't have a negative voltage source for the OP, instead I could have a constant voltage at the output, and thus current flowing though the earphone, so the signal could move up and down, but without going below 0, simulating AC, but I guess this would limit the variation of voltage (and therefore volume) before exceeding the power rating. Is there any way to transform this "simulated AC" into real AC, maybe a 1:1 transformer?

Pic related, shitty OC

[Anonymous No.806239]

>>806235
Yes I was using a 33k resistor. I swapped it to a 22k resistor to make a numeral cathode glow brighter and accidentally connected it to a comma one.

[Anonymous No.806253]

>>806239
Maybe the comma needs a smaller current since the surface area is smaller, therefore a bigger resistor. Do you know if there's a datasheet about that nixie? Take into account that 150 volts flowing through a 22k resistor equals 1W, so expect to burn the typical 1/4 watt resistors if you use them.
>>806235
Respect to this, I've made a circuit simulation, I biased an OP amp and added a capacitor at the output so there's no constant current flowing though the earphones, the current should only pass when the waveform changes. I guess this will be very similar to the jacks of computers and phones, the "simulated AC" get's converted into "true AC" thanks to that cap, right?

[Anonymous No.806254]

>>806253
Pic related

[Anonymous No.806266]

>>803553
Thanks for the links, looks useful.

Have you tried making any relatively fine-pitch boards? I mill my own boards too, and 0.5mm pitch stuff has been a real problem for anything more that a couple of inches square.

Pic is the result of laser exposing then etching, which seems to work better than milling for the fine stuff.

[Anonymous No.806319]

Just checked simulating my earphones connected. It's just what I wanted. I'll add a divider so I can adjust the volume.
Last question I plan to use an LM741 as the OP amp to amplify a 0-20mV signal (or at least that's what I think a mic outputs). Is a 101 gain good in terms of noise and response? Or should I use two OP amps to have 11 gain each?

[Anonymous No.806336]

>>806319
Couple of things:
- LM741's maximum output current is typically just 25mA, so you can't expect 2V output to a 16R load.
- The gain-bandwidth product is around 1MHz, so a single amp with G=101 won't cover the whole audio range. You need two amps.
- The input and output voltages are limited couple of volts away from both rails. This can be a problem with 9V batteries, particularly with rechargeables.
- The noise advantage of a G=101 amplifier is negligible over 2 x G=11 amps. Use better opamps, be careful with layout, use small resistances, etc. if you want lower noise.

[Anonymous No.806338]

>>806336
I'll solve the current problem with a transistor
I'll use two OPs then.
The output voltage doesn't needs to be too high, it's just to drive some earphones.
I'll better use two amps so I can increase the gain even further if I want to.

[Anonymous No.806428]

>>806336
Should I set the two OP as inverters to end with the same signal? Or two as non inverters? Does it matters in terms of noise and overall performance?

[Anonymous No.806435]

>>806428
>Should I set the two OP as inverters to end with the same signal?

definitely, otherwise the waveforms will be flipped and it will only sound right if you're standing on your head, or you move to Australia.

[Anonymous No.806436]

>>806435
But if I set them as non inverting I end up with the same signal as if I set both to inverting (plus +1 gain each)

>signal- Not inverted -signal- not inverted : signal not inverted

>signal- inverted - inverted signal - inverted again: signal not inverted

In both cases I end up with a non inverted signal.

[Anonymous No.806441]

>>806436
Wut, these are supposed to be inverters, why the first one is acting as a non inverter?

[Anonymous No.806446]

>>806436

doesnt make any difference which polarity the signal comes out as. people cant tell the difference, only instruments can.

[Anonymous No.806455]

>>806446
I knew that, I just want to stick to what I'm trying to make, I've made a thread about OP amps to stop bloating this thread, there are still some problems that I haven't solved yet.
I guess it could also be helpful for people who want to learn more about OP amps.
>>806452

[Anonymous No.806469]

>>806428
Excuse me real quick.
Have 555 timers become a 'meme' on this board??
I haven't been here in a while...

[Anonymous No.806474]

>>806469
It's not like the "watch boku no piko" meme. When people suggest using a 555 they mean it. A 555 based LED flasher was the hello world of electronics before Arduino took over.

[Anonymous No.806480]

Hey /diy/ I'm looking for a 32-6 volt smps for driving high powered leds, however everything I see in that range is something cheap, I'd like one with a couple hundred watts and around 90% efficiency. Basically a good pc power-supply but 32v instead of 12v.
I'd like to build my own but one of such a size is beyond my skill level. Any suggestions?

[Anonymous No.806484]

>>806469
The 555 has always been something of a meme.

Most people (before an MCU dev board was $15) built a 555 blinky light before they even knew how a transistor worked. The 555 is also somewhat revered because it was one of the first ICs ever, and is still really widely used. This goes way, way further back than /diy/ or 4chan existing.

Just for fun:
http://www.evilmadscientist.com/2013/555-kit/

[Anonymous No.806489]

>>806484
>>806474
Oh yeah i'm familiar with the 555, I'm an electrical engineering student. I just haven't seen it talked about so much here. I browsed back before there was an /ohm/ or whatever

[Anonymous No.806509]

>>806480
>I'm looking for a 32-6 volt smps for driving high powered leds,

a largish laptop power supply is 18V @5A. stick two in series and you get 36V @5A, or about 180W. you can get 'em for $3 each at a thrift store, so it involves very little cost, and almost no work. like all projects should!

[Anonymous No.806545]

>>806428
"Always invert, except when you can't. And usually you can."
Opamps generally perform bit better when operating in inverting configuration, but IMO you should just do what you find more convenient and intuitive.

>>806484
Yeah, at one point it was an equivalent to an Arduino LED blinker - people used a whole IC with dozens of components to do something two transistors or even one transistor could do.

[Anonymous No.806602]

>>805591
Who is the cock smeg forcing everyone to slap his hand?

[Anonymous No.806632]

Guys, I have a problem:
My blender have this switch KW4AS (picture related) But it won't stop making sparks and melting the plastic of the switch, the blender can't be turned of and have to be unplugged.
I want to replace it (if you know where I can find the same let me know) but I want to know how the hell I can clean and fix a switch like that. Have you any idea of how it became a permanent short circuit?

[Anonymous No.806657]

bump test

[Anonymous No.806666]

>>806469
All praise the mighty 555
>>806545
>Opamps generally perform bit better when operating in inverting configuration
Thanks for the tip
>>806632
Better buy a new one, they're called micro switches, make sure you buy the right rating. Vibrations from the blender could cause bad contact, creating sparks and overheating.