/diy/ - Do It Yourself

Rest in pieces of smashed lame circuitboards, fellow alcoholic :)

>name
>value

every time man, every time

>>987082
"He installed "hassler" devices that emitted high-pitched sounds whenever someone talked too loud, and even blew up an annoyingpublic addressspeaker with firecrackers.[11]Jim Williamsrecalled an incident when, after tracing externalelectromagnetic interferenceto thecontrol towerof theSan Jose airport, Widlar telephoned the airport and demanded they shut down the transmitter"

What a fucking legend.

i hate digikey so much there has to be something better for canada fags..... right? :(

>>987270
jesus christ that spacing

who /operator/ here

I bought a cheap HDMI splitter that is very popular, but its known that it runs pretty hot.
The metal case was warm to the touch after about 5 minutes of using it.

After only running for 5 minutes the small chip on the left was running at 70c, the two bigger ones were running at like 60c. This is without the case on and only a short time.

I am going to 3d print a new top of the case for it with a 40mm fan blowing down on top of it, 3 sides of it having air vents.

I have mini heat sinks that came with some stepper motor drivers I upgraded that will fit on top of chips. Should I put them on?

I have read somewhere that adding small heatsinks to a raspberry pi will actually make it run hotter.
Is this horseshit or will they be beneficial?

When I use this, its gonna be running for 16 hours straight.

>>987082
He invented the Widlar current source. Useful for IC's that might need bias currents that are too small for a standard current mirror to produce

>>987413
>Should I put [heatsinks] on?

Yes. Before you do anything else just add heatsinks. That'll probably make enough difference alone that you won't need to do anything else.

>>987413
Don`t forget to put a bit of copper toothpaste between your sinks and chips.

>>987108
TFW our company is still riding the eagle.
Muh Altium D.

>>987613
I used thermally conductive double sided Kapton tape.

>>987534
The temps dropped down to about 40c with the heatsinks and a small fan.
I wonder why they wouldnt have just added heatsinks in the first place.

You see a lot of people say it works great but died after x amount of hours, and I would guess is due to how long they actually let it cook itself to death.
A feat heatsinks cant be more expensive than their return rate

I have a question about "ground" in electronics. In electronics earth is commonly used as a reference point; it is said that earth has a common potential.
In circuits two nodes have always the same potential only it there is no resistance between them, which makes them basically the same node, or a short if you prefer to see it that way.
How can this be possible with earth?
Lets say we have two points on the surface of the earth that are 10 km away. Wouldn't the soil have a significant resistance between those points?

>>987845
It does, but the assumption is that the earth is a large enough body that it can easily absorb or discharge charge without noticeable voltage shifting. Imagine that the earth is a massive capacitor, that's basically how we can treat it as a 'ground".

>>987845
> Lets say we have two points on the surface of the
earth that are 10 km away. Wouldn't the soil have a significant resistance between those points?

With AC, it's not necessary for electrons to be able to travel all the way around a circuit. Because the current reverses periodically, the total amount of charge involved is fixed. So long as you can get a certain amount of charge into or out of each electrode, current will flow. This is why AC can flow through a capacitor (two conducting plates separated by an insulator).

Furthermore, any conductor has some self-capacitance. Any lump of matter with the same number of protons and electrons is at ground potential. An imbalance between the two results in a force which will act to correct the imbalance. But this force is finite; if you have an electromotive force (voltage), you can use that to force electrons into or out of a conductor against the potential force.

Consequently, it wouldn't matter if you insulated the two halves of the planet with a perfect insulator. You can force electrons into one half and out of the other without any need for current to actually flow between the two halves.

And you don't even need a particularly high voltage to do it. Dirt may be a very poor conductor (i.e. it doesn't have many free electrons per unit volume), but there's just so much of it.

>>987845
Do not confuse circuit ground with your house's earth/ground circuit.

Circuit ground is whatever you make it up as on your circuit. It doesn't necessarily need to be your house's earth.

your house's earth is (usually, in the US) bonded to neutral from the power company at your meter.

You usually also have a ground rod in place there. If you pound a rod in 20' to the side, and measure the resistance between you house and the new ground rod, it will be thousands of ohms.

If you connected a live leg of your house to the new ground rod, it'd happily dump a few watts into the ground rod forever, and NEVER trip your circuit breaker. You'd just have a hot wire going into the earth.

Now if you ran a #6 copper wire between your two ground rods, you would improve your house's ground circuit, and THEN you'd have you 120v line going through like a 5 ohm or less circuit, then it'll happily trip like it's supposed to.

you can also have several volts just between different spots around your house.

>>987845
>Lets say we have two points on the surface of the earth that are 10 km away. Wouldn't the soil have a significant resistance between those points?
MOST of the physical Earth (the dirt we are standing on) has enough moisture in it that it can absorb huge amounts of local charges--so essentially, the whole thing is considered to be at the same electrical potential even though it is not.

In locales with very rocky and dry soil, the "grounding" ability of the soil may be so poor that special steps must be taken.
The main time I've heard of this issue is with properly grounding radio transmitter antennas. There is info online about shortwave antenna grounding issues like this.
In extreme circumstances when good grounding is not possible, a special type of ground plane must be added on to the transmitting antenna:
https://en.wikipedia.org/wiki/Counterpoise_%28ground_system%29

An antennas ability to radiate EM waves depends partly on it having a good ground connection--and the transmitter circuit may be damaged if it cannot radiate power from the antenna properly.
With big transmitters, you can't just string a piece of wire up a tree if you don't have a real antenna. The antenna+ground has to be a specific (wave)length and orientation or your transmitter is gonna overheat and burn up.

>>987845
>I have a question about "ground" in electronics. In electronics earth is commonly used as a reference point; it is said that earth has a common potential.
>In circuits two nodes have always the same potential only it there is no resistance between them, which makes them basically the same node, or a short if you prefer to see it that way.
Also always remember: SPEAKER "grounds" are NOT the real ground.

In a lot of schematics, [+] is the positive DC power and [-] just connects to ground.
So you might come to think that anything marked with the negative sign can be connected to circuit ground--and for everything else that is usually true--but for speakers that is generally incorrect.

Speakers use the [+] and [-] symbols to represent the two different wire connections to a speaker--but even though one is marked [-], it isn't ever connected to the circuit ground. Stuff will burn up instantly if you do that. Don't ever connect the speaker negative to circuit ground unless the schematic tells you to.

Sometimes they refer to the speaker negative wire as a "floating ground".

goddamn bob pease said that shit is true

>Widlar's eccentric behavior is remembered in legends and anecdotes which, according to Bob Pease, are largely true.

>However, the story about Widlar bringing a goat to trim the lawn in front of his office, retold by The New York Times after his death,[8] was incorrect.[11] It was a sheep, not a goat;[71] Widlar brought her in his Mercedes-Benz convertible for just one day, which included a photo op for the local journalists.[11] According to Pease, Widlar abandoned her in the nearest bar;[11] according to Lojek the sheep was "mysteriously stolen".

>>988103
>"grounding" ability of the soil may be so poor that special steps must be taken
Yeah, that applies to your grounding stakes or whatever you're using to make a contact with soil. Grounding resistances can vary from ohms to hundreds of ohms. The actual ground resistance/impedance after that is very low and it isn't some "big capacitor" thing which only works for AC. For example, soil is occasionally used to replace the other wire in high voltage DC links in a similar manner as in SWER systems.

How do I find a transfer function for any given circuit?

Assume I know basic circuit theory and have had math up through differential equations but I am rusty. Also assume I've not taken any analog signal processing courses and have no other experience with LTI systems.

In doing some very quick reading it seems like a lot of nodal analysis (which I have experience with) in the s-domain (which I don't even know what the fuck that is.)

I'm trying to get the transfer function for a filter I'm using in a project. I can find it online but I'd prefer to derive it myself. I'm hoping I'll gain a more intuitive understanding of exactly how it works by doing so rather than pretending that filter is a magical black box that attenuates the signals I don't want passed with the power of magic smoke.

Thanks for all of the replies, I really appreciate the effort.

>>987847
So in this analogy the earth is a capacitor so big, it can be charged almost infinitely without noticeably changing its voltage? How big a capacitor like that would need to be?

>>987857
>Any lump of matter with the same number of protons and electrons is at ground potential.
Is that also true with the Earth (as in the planet)? Isn't it at least slightly positively or negatively charged?

>Dirt may be a very poor conductor (i.e. it doesn't have many free electrons per unit volume), but there's just so much of it.
Do you mean it's like with connecting many high resistances in parallel, which results in a low total resistance?

>>987860
>Do not confuse circuit ground with your house's earth/ground circuit.
I know that ground in circuits can be any node you pick. I just don't know how can we use the (literal) Earth as a reference voltage, a node which has the same voltage everywhere.

>your house's earth is (...) bonded to neutral from the power company at your meter.
>You usually also have a ground rod in place there
What's the point of putting the rod into the ground there if you have a neutral wire going to your house?

>If you pound a rod in 20' to the side, and measure the resistance between you house and the new ground rod, it will be thousands of ohms.
If the resistance is so high, I fail to see the point in involving the physical earth at all.

(continued in the next post)

>>988103
>MOST of the physical Earth (the dirt we are standing on) has enough moisture in it that it can absorb huge amounts of local charges--so essentially, the whole thing is considered to be at the same electrical potential even though it is not.
I'd like to see a circuit diagram of that. Does it have something to do with the capacitor analogy of >>987847 ?

>>988112
>Also always remember: SPEAKER "grounds" are NOT the real ground.
So you mean there can be a situation in which the speaker negative terminal "should" theoretically have the same voltage as the real ground, but isn't connected to it directly? Why would it blow up if I connected it to the real ground?

>>988178
So the Earth resistance is actually small?

I generally have a hard time imagining how can the Earth have the same potential everywhere, and not being a perfect conductor at the same time.

Maybe you could try to draw me a diagram of a circuit equivalent of a ground connection between two points at some distance? I'm interested in what order of magnitude is the resistance between them (or capacitance, referring to >>987847), what it depends on (like moisture, obviously distance, etc.). That would make things clear for me.

Sorry if my questions seem dumb, and thanks in advance for all of the answers!

>>988300
>You usually also have a ground rod in place there
>What's the point of putting the rod into the ground there if you have a neutral wire going to your house?
The wire from the transformer to your house may be several hundred feet long, the voltage drop across the neutral conductor would be significant. Both ends get grounded for safety incase high voltage (>500v) gets on it.

>If you pound a rod in 20' to the side, and measure the resistance between you house and the new ground rod, it will be thousands of ohms.
If the resistance is so high, I fail to see the point in involving the physical earth at all.

Either one of the grounds might be a couple volts apart, and thousands of ohms between them. If you bond them together with a low resistance wire, the resistance goes to nearly 0, better than either the wire connecting them, or the earth alone.

It's the difference between 105v in your house or 120v. The hot lines coming in from the power company actually haven't changed, but your neutral/ground got better.

>>988302
There are several things in effect. For radio stations to work, they might need a good local ground for their antenna (in which case earth forms a part of the antenna system), but there's no need for the transmitter and the receiver to share their ground, as energy is transferred as radio waves. Cell phones, keyfob transmitters etc. wouldn't work if a common ground was required.
Then there's that big capacitor thing (Wikipedia suggests the capacitance of Earth is around 710uF). Afaik that thinking comes from the olden times and electrostatics. Whatever charges man-made gadgets are capable of injecting to earth are negligible and are countered by the generated opposite charge, so the total charging state of Earth can be assumed to remain constant.
Then there are systems which really use earth as a conductor. The applications are quite limited, but saving one wire in (some) power distribution system is a very important use. You can google something like "SWER earth resistance" and get decent articles about the topic. Those articles suggest values like 0.5 ohms per kilometer and below for normal cases and inductances in order of 1mH per kilometer. At low frequencies and DC the ground currents go pretty deep, 1km or so. However, you need a contact with earth and typical (man-made) grounding connections have relatively high resistances, measured ohms in good cases and in hundreds of ohms in bad cases. In other words, the ground connections itself often dominate the total resistance between two earth-grounded points.

>>988262
>s-domain
Long story is, you do a Laplace transform to the s domain (where s=j*omega) and then it's simple complex algebra.

Short story is, replace capacitors with 1/(j omega C), replace inductors with (j omega L), solve as usual. Take the absolute value to get a function in terms of omega.

If that didn't help, I'm sure people will help you derive your transfer function in this thread if you're clearly putting in the effort.

I was playing with a nigger-rigged electromagnet, just an enameled copper wire wrapped around a thin bolt.
The whole thing runs to a 12V power supply.
When I turn the NPN transistor on via the 22K resistor, the MOSFET gets too hot to touch in few seconds.
Brushless motor controllers basically turn electromagnets on and off, and they get warm at most...so what am I doing wrong?

The MOSFET I'm using is a beefy IRF9530

>>988262
> How do I find a transfer function for any given circuit?

> it seems like a lot of nodal analysis (which I have experience with) in the s-domain (which I don't even know what the fuck that is.)

Nodal analysis gives you a bunch of equations relating the currents from Kirchhoff's current law and a bunch of equations relating the voltages from Kirchhoff's voltage law.

If the only circuit elements were resistors, you'd be able to convert between currents and voltage using Ohm's law. This would give you a set of simultaneous linear equations which can easily be solved for the voltages and currents. Substitution gives you ΔVout/ΔVin (the transfer function, which will be a constant for a linear system).

When capacitors and/or inductors are involved, Ohm's law is insufficient; you also need I=C.dV/dt and V=L.dI/dt. So your system of linear equations becomes a system of differential equations.

This is where the Laplace transform and the "s domain" comes in; it allows you to transform differential equations into algebraic equations (which in this case will be linear).

Having done so, you can solve the equations to get Vout/Vin as a function of s, which is the transfer function.

>>988419
Here's how it looks like on a breadboard.
The electromagnet works

>>988419
Connect the source to the positive rail and the drain to the coil.

As it stands, you can't turn the FET off. If Vg=V+, it will conduct until Vs rises to the point where the FET is in the linear region.

You always put the load on the drain so that Vs is fixed and Vgs depends only upon Vg, and not on the voltage drop across the load.

This is true for both N-channel and P-channel. Normally N-channel is preferred (for "equivalent" parts, the N-channel version has better specs and/or is cheaper), but if you need one side of the load to be grounded, then you use P-channel precisely because this lets you connect the load between drain and ground.

>>988426
Sorry, I maybe got the diagram wrong.
The source is connected to Vcc and ground to coil. It works, but the mosfet gets burning hot. I guess there is too much current flowing through it - the coil is basically a short circuit. So should I somehow limit the current? How do motor controllers deal with this?

>>988429
*drain is connected to coil

>>988302
There is no "circuit". A lump of matter acts like a capacitor with one side connected to ground. You can push electrons into it or pull them out of it.

The greater the voltage, the more electrons you can move before reaching equilibrium (where the force arising from the charge imbalance equals the force from the applied voltage). I.e. Q=C*V.

>>988429
The simplest way is by not using too high a supply voltage, so that the peak current (supply voltage divided by resistance) isn't too high.

The other option is to use PWM, effectively forming a buck converter which regulates coil current rather than output voltage. I.e. you measure the coil current and use a negative-feedback loop to turn the FET on and off in order to maintain the desired current. The comparison needs some amount of hysteresis to avoid too high a switching frequency or (worse) operating the FET in the linear region.

>>988354
>>988420
Thanks for the advice. I've done Laplace transforms before but I'll need to review it a bit. It sounds like I was kind of on the right track.

The only thing I'm not totally sure of is how to handle components like diodes and active components like transistors and op-amps. When I learned nodal analysis it was for RLC circuits. I'm trying to get the transfer function for a multiple feedback bandpass filter and I'm not sure how to treat the node between C1 and R2 and the the node on the output.

My guess for the former would be:
V(n3) = 0V because virtual ground therefore
[0 - V(n2)]/(1/sC2) + [0 - V(n4)]/R2 = 0
and for the latter I'm less sure because I don't know how to treat the output of the op-amp as a node. Assuming current is flowing out of the node in every direction I have something like:
[V(n4) - V(n3)]/R2 + [V(n4) - V(n2)]/(1/sC1) + I(xop1) = 0

>>988450
> The only thing I'm not totally sure of is how to handle components like diodes and active components like transistors and op-amps.
Diodes are problematic because of their inherent non-linearity. Transistors and op-amps aren't as a problem so long as their behaviour is linear; they just add more equations to the system.

> When I learned nodal analysis it was for RLC circuits. I'm trying to get the transfer function for a multiple feedback bandpass filter and I'm not sure how to treat the node between C1 and R2 and the the node on the output.
Virtual ground => V3=0. KCL at node 3 => V4/R2 + C2.dV2/dt=0.

> and for the latter I'm less sure because I don't know how to treat the output of the op-amp as a node.
The op-amp's output is a voltage source (with the other end at ground). You can use either the op-amp's transfer function (V4=-gain*V3); or the virtual-ground equation V3=0 (which is just the former equation when the gain is infinite).

>>988450
In regards to diodes and FETs, you might want to learn about small signal models first.

>>988419
>When I turn the NPN transistor on via the 22K resistor, the MOSFET gets too hot to touch in few seconds.
>Brushless motor controllers basically turn electromagnets on and off, and they get warm at most...so what am I doing wrong?
Nothing is wrong. Coils fed DC just get hot.
Coils don't have significant resistance, just inductance--which is resistance to a changing current flow. If you apply DC the current flow stabilizes and the coil only has the actual wire resistance left to limit current.

This is why stepper motors left enabled but not moving tend to get way hotter than when they're in use.
And it's also why some stepper drives have the option of a lower "holding" current they can switch to after a second or two of not moving.

What you could do is just build a circuit to switch the coil on and off pretty fast--1000 times a second or whatever.
Wall-wart transformers are okay at 50/60 Hz, but they have a lot more inductance (way more windings) than your puny nail coil.

>>988437

>>988450
Stuff like diodes and transistors in the feedback loop, I would just SPICE honestly. Though a diode isn't that bad in the DC case.

Who among us has done anything with ESP8266 modules?
I haven't done anything yet, finally have what may be a use for them so I ordered a few different kinds (pic related, some but not all of them).... I may have ordered what amounts to basically duplicate items, but they were cheap.
It'll be a couple weeks at least until I get any.
Is there any forums that are good for these, other than the esp8266.com one?
Is there a (non-Arduino-but-still-fairly-easy-to-use) IDE that lets you program them in Lua? The only examples I've seen just people keying stuff into a terminal window, in real time.... I'm not real sure how useful that would be.

So I'm in the early stages of design for a uni project.
It's basically going to be a 4-wheeled robot that will drive around farm land identifying plants and maybe animals.
For starters, I need electric motors and batteries to get it moving. It'll need to drive around for at least 4-6 hours a day, charge at night etc.
This is my first big project, so I've been browsing for parts and I found this: https://www.motiondynamics.com.au/brushless-gear-motor-60v-1500w-600-700-rpm-60nm-controller-accessories.html
I think it would be a good motor because it's small, light and seems to have enough power. But now I need to find some sort of battery. How many AH should the battery have? Where can I find prices and such?

I'm also planning on powering everything with an Intel Edison, since I assume I'll need a bit of power for the machine vision I'm planning on implementing.
Any other general tips would be greatly appreciated since this is my first major project.
Thanks guys.

>>987082
hey /ohm/
Trying to get my head around basic electricity, and later re-school (in a way) from joiner to electrician.
Anyone else done this, who would like to share experience or advice?

I have a question which my textbook doesn't explain, and was hoping someone here would.
-If electrical current flows from the negative side of a battery to the positive side, why do every illustration i have of DC show a positive charge?

(doesn't these electrons repel from same-charge item? i might have gotten all of this wrong)
And why does people teach as if the current was flowing from + to -?

sorry if this is stupid question, but i cant get my head around it, and for every example i read in the book i just get more confused.

>>988812
So it's basically a machine vision project, except for some reason you're going to build the robot yourself?

Power consumption is fairly hard to calculate. At low speeds (before air resistance starts to matter), a given terrain will require a given amount of energy per unit mass per unit distance (i.e. it will be as if you're continuously climbing a slope of a given angle). Smooth, flat terrain will require very little energy, rough terrain a lot more. You'd probably need to find examples of similar projects and just scale accordingly.

The main factor you've omitted to mention is speed. You can run it for 4-6 hours on a AAA cell provided you're only moving at 1mm/sec. You should be specifying distance, not time.

Note that if the mass starts to be dominated by the batteries, using more or larger batteries will just increase the energy requirements in direct proportion to energy availability, i.e. it doesn't get you anywhere.

>>988840
> -If electrical current flows from the negative side of a battery to the positive side, why do every illustration i have of DC show a positive charge?
"Current" flows from positive to negative. Electrons are negatively charged, so a flow of electrons from the -ve terminal to the +ve terminal corresponds to a flow of current from the +ve terminal to the -ve terminal.

In a solid conductor, the nuclei (and thus the protons) are held fixed, and only the electrons move, thus all current flow corresponds to movement of electrons in the opposite direction to the current. This isn't the case in a gas or liquid, where the current also includes positively-charged ions flowing from +ve to -ve.

Electronics noob here, what powersuply can i use to replace 2 AA battreies in a device?

>>988993
3V DC. Depending upon the device, a 3.3V supply may be close enough.

You also need to check the current rating. If you don't know the current draw, time how long it takes to drain a pair of new cells with a known capacity. Current (mA) = capacity (mAh) / discharge time (hours). Note that the capacity is that for a single cell (putting two cells in series doubles the voltage, not the capacity).

>>988993
>>988997
>3V DC

If they're in series. Which they almost certainly are. But ya know.

>>988908
>So it's basically a machine vision project, except for some reason you're going to build the robot yourself?
I'm not sure what you mean by this. Of course I'm going to build it myself, it's a uni project.
As for speed, it isn't too important. I imagine it would go at a walking pace or maybe a bit faster.

Also, the motor is 60v and rated for 32A. Does this mean I need a battery that can output 60v minimum? Or can I use a transformer to step up the voltage or something?

Hi /Ω/

I have a bunch of CD 4017 counter ICs that I'd like to use in some little project.

I'd just like to hear some ideas about these ICs and how they could be used.

I was thinking of using a microcontroller to drive the CD4017 at a high frequency (250 Hz or so). Each of the outputs would then switch on/off at 25 Hz. Maybe these outputs could be used to make a simple PWM controller?

>>989089
>a simple PWM controller?
Your microcontroller does that much better already.
I've seen those used in roulettes, "analog" clocks, simple sequencers (in music stuff), simple event counters with dot displays and such.

You're trying to do things backwards, though. It doesn't make much sense to try to forcibly find uses for some random $0.10 IC. Even less, when you need more expensive ICs like microcontrollers to pander them.

>>989075
> Also, the motor is 60v and rated for 32A. Does this mean I need a battery that can output 60v minimum?
Those are both maximum values. And unless the robot is going to be lugging around a *lot* of mass, the motor is almost certainly overpowered. E.g. in countries where you can ride a "motorised bicycle" without a licence, 1.5kW would typically exceed the maximum allowed power. That motor is more than capable of lugging an adult around at 30 mph.

> Or can I use a transformer to step up the voltage or something?
I suspect that this project is over your head. DC-DC power conversion is fairly simple compared to most of the other stuff involved. You say it's a "uni project", but clearly you're not studying EE (or if you are, you've been asleep the whole time).

If you're coming from a CS background, I'd strongly suggest buying an off-the-shelf robot and just doing the computer vision aspect as your project.

>>989089
> I have a bunch of CD 4017 counter ICs that I'd like to use in some little project.
> I'd just like to hear some ideas about these ICs and how they could be used.
The stereotypical use for decoded counters is LED chasers (think "Knight Rider", if you're old enough to have seen that).

If you're using a uC, you wouldn't bother with the counter, you'd just drive a 3-to-8 or 4-to-16 decoder from 3 or 4 GPIO pins and let the uC do the counting. Or just use 8/10 GPIOs if you have them spare.

Small-scale (e.g. 74- or 4000-series) logic is far less useful than it used to be before uCs became viable for hobby projects.

>>987082
>Each of the outputs would then switch on/off at 25 Hz. Maybe these outputs could be used to make a simple PWM controller?

With PWM the thing you want to control is the duty cycle (width of high pulse relative to low pulse). With the 4017 decade counter each of your outputs are on 1/10th of the time as it cycles. Your pulse width is always going to be 10% no matter what the frequency is.

>I'd just like to hear some ideas about these ICs and how they could be used.
pic related

>>989089
>high frequency
>250 Hz

>>989089
>PWM controller
to be pwm you need to be able to alter the width of the pulses. you can't independantly control the mark/space ratio or frequency with a setup like this. maybe if you had a shift register or something.
usual thing to use them for is a clock, connect each output to an led representing the position of a hand, 60 led's for minute hand etc.
i don't know how about one of those strong man hammer hit fairground things.

>>988840
Goes back to Benjamin Franklin and the difference between conventional current vs. electron flow.

https://en.wikipedia.org/wiki/Electric_current#Current

>>988809
Yeah I just built a door monitor with one. One of the implementations of the software I did was using that Lua nodemcu firmware. I used ESPlorer to load the scripts onto the board. Its OK but seems kinda buggy. Pic related is ESPlorer but its not difficult to roll your own software to at least load Lua scripts onto the device.

>>989156
I'm glad Franklin made that "mistake". If he didn't most of us would probably assume that the current flow is the same as the direction charges travel. But thanks to it now we have to explicitly learn that current flow is just a convention and as long as we stick to it it doesn't matter which way charges actually travel.

Does anyone here have experience working with Nitinol (Flexinol/Muscle) wire?

https://www.sparkfun.com/products/12096

I want to pull back a small rod inside a "tunnel" that is only 0.5 inches in diameter as fast as possible and as frequently as possible. The rod must travel 20mm.

Using a larger solenoid outside the tunnel is not possible since there is not enough space.

Would Nitinol be able to accomplish this, or am I asking for too much?

>>989701
is this because you want to add a vibrating knob to your /diy/ sex robot?

>>988911
>>989156

Thanks guys, appreciate it.

>>987688
>I wonder why they wouldnt have just added heatsinks in the first place.

Because you're just a wuss. 70 degrees Celsius is not a bad temperature at all for small high-speed electronics to run at, as even the cheapest electronic parts will have to be specced at 85 degrees Celsius min. Also when the IC's will be dead because of long exposure to that temperature the device will be so obsolete that there will have been 3 successors to HDMI. So in fact you've just wasted some copper or aluminium for no purpose at all.
But I guess you're no engineer and never will be right? right?

>>988809
>Is there a (non-Arduino-but-still-fairly-easy-to-use) IDE
Why not just use the Arduino IDE?
There's an ESP8266 core for it that works fine.

I like it much more than using Lua.

>going to take my final amp and shit exam this morning
Took two semesters to learn how to understand a 741.

COLLAGE

>>989850
>run the processor at high temps
>put it in a box
>cram it in with other chinky components that will heat up
>trust it all to work flawlessly

>>989850
>Being this butt-devastated over heat sinks.
Your parents get raped to death by some aluminum fins?

newfig here.

Id like build a variable metronome that pulses different color lights for different beats. I can program a microcontroller no problem, but im not sure what components i nees to control it with.

Initial ideas are:
Pot for variable tempo control
Two knob selectors (from like 1-10 points) to determine first and second measure beats for mixed time options. Not sure what these are called or if there is a better way to choose from about a dozen options.

Also any microcontroller recs? Only needs to handle 2 led outputs and 3 inputs, although 5 inputs woyld be nice

>>991077
> Two knob selectors (from like 1-10 points)
Binary switches would work well here. It's a switch with multiple positions that mechanically produces a binary output. Pictured here are the thumb wheel variety, there's also rotary ones.

You could use a regular 10 position rotary switch and convert that to binary with a priority encoder or just read all 10 outputs directly into the microcontroller if there's enough pins.

Rotary encoders are also an option, but you need to keep track of their position as a variable in the microcontroller because they just produce a pulse when turned clockwise and a different when turned counter clockwise.

Cheers anon

Why was the transistor such a huge step forward for computing machines?
I mean, I see all of the advantages over previous technologies, but why wasn't it possible to build logic-level stuff with, let's say, relays? Is it just because chips are just a fuckload of components put together and would take up several rooms at relay scale?

>>991112
>why wasn't it possible to build logic-level stuff with, let's say, relays?
But it was and such computers were made. They were very slow, though.
It was possible with tubes as well. Several tube computers were made.

The problems with both were size, enormous power dissipation and unreliability which also limited how complex computers you could build. Transistors solved those problems and also got faster than tubes in computer applications pretty quickly.

Have a video: https://www.youtube.com/watch?v=MM4IUxXiArY

>>991112
There were logic circuits made with relays used in railroads and other applications. The problems with relays isn't just their size: A typical relay is only good for 100,000 cycles before the mechanical parts break down. They are also very slow; around 0.1 seconds to make the switch. The Commodore 64 was 1MHz. if it's logic were implemented in relays it would be 10 Hz. Something that could be done in a second on the C64's transistor logic would take a day with relays, and you'd need to stop everything and replace a relay every few minutes. Releys also take a lot more power and generate a lot more heat.

>>991112
> why wasn't it possible to build logic-level stuff with, let's say, relays?
It was. You could even build a computer from hydraulic or pneumatic valves if you had a need to avoid electricity altogether.

> Is it just because chips are just a fuckload of components put together and would take up several rooms at relay scale?
If you tried to make a PC with relays, it would take up several large buildings, require its own power station, would run 100,000 times slower, and probably wouldn't even finish the boot sequence before one of the relays failed.

Relays were replaced with valves (tubes) due to speed and reliability, valves were replaced with discrete transistors then ICs due to size, power consumption, and cost.

>>991154
> would run 100,000 times slower
Make that 100 million times slower (tens of Hz, compared to GHz).

I'm trying to troubleshoot a pretty basic power supply circuit inside of 4 track cassette recorder. I tried measuring the difference between input and output voltage at the 317 regulator but both are zero. So I'm trying to test the DC jack but its only reading 1.2v DC whereas the male bit coming from the plug reads in around 18v (12v supply, assuming because it's not under load its measuring higher?)

Does this mean I just have a bad jack? The solder points look a little frosty but how the fuck can a simple power jack fail?

>>991077
>Id like build a variable metronome that pulses different color lights for different beats. I can program a microcontroller no problem, but im not sure what components i nees to control it with.
>Initial ideas are:
>Pot for variable tempo control
>Two knob selectors (from like 1-10 points) to determine first and second measure beats for mixed time options. Not sure what these are called or if there is a better way to choose from about a dozen options.
Warning: opinions ahead...

The obvious choice for a 10-way selector is you get a 10-position rotary switch, and solder some 1K resistors between all but one of the pairs, and then have one more hanging off. This gives you a variable resistor knob with 10 "clickable" positions that you can read with an analog-in pin. BUT,,,, cheap rotary switches wear out if used a lot. And good rotary switches cost $$$.....

If you use capacitive buttons to control the whole thing, they won't wear out. You can make them, but the modules are cheap online too.

If you MUST have knobs,,,,, then you can use a stepper motor as a rotary encoder also, and it isn't gonna wear out. A normal 200-step motor has 50 detents per turn.

You'd want a LCD on there to show the numeric value directly, since you wouldn't be able to just read it off the controls.

I would use ws2812 / neopixel LEDs since they already allow adjustable color and brightness, and you can run all of them off 1 digital pin.

For what board to use, I am lazy and usually just get a Mega. Get a 12v power supply and a buck converter, and feed 7 volts to the Mega.

>making this stereo volume knob
https://www.youtube.com/watch?v=sJ5vhShdVjo

I have mine made, but I just get regular audio turning it one direction and a glitched mess in the other. I'm pretty sure I have it wired correctly.

He connects the ground to two pins of the potentiometer then bridges them, wouldn't you just have the ground go straight through and not touch the potentiometer at all?

>>991154
There are working examples of relay and fluidic computers.

>why do computers need to be small
Why do roads need to be short?
https://en.m.wikipedia.org/wiki/Grace_Hopper

Hi /ohm/ can you help me out? I can't obtain the equation that determines the charge time of the capacitor and thus the frequency of operation depending on the resistor at the top of the circuit.

>>988997
>>989067

Its... for a generic vibrating bullet used for sexual stimulation

Can a keyboard (musical) matrix be done with an Arduino?

I have a guitar effects pedal that no longer powers up. The battery and wall plug wires have continuity into the circuit board so it seems like a component I recall using a higher amperage transformer than was recommended in the past, so perhaps that fried something. Problem is i dont knwow aht to look for. All the solder joints look fine, i dont have a schematic (propriety bullshit) and theres just a ton of stuff that i dont know if it should be series/parallel whatever. Any ideas of what to check? Ill be trying everything for continuity but its gonna take a while

Tldr; what type of component is most likely to fail when a circuit is overpowered?

>>991411
yes

>>991411
No.

>>991398
The time constant of a R-C filter is R*C. So for R=10k and C=2.7nF, T=27us.

That's the time taken for the difference between the output voltage and the input (equilibrium) voltage to decay by a factor of e (=2.718...).

The voltage divider for the positive feedback has a 10/11 ratio, so the voltage difference has to decay by a factor of (11+10)/(11-10)=21. ln(21)=~3.0445 Np, so the interval between transitions is ~= 82.2us, the period is 164.4us and the frequency 6083 Hz.

>>991443
The PSU's current rating doesn't matter. But if the circuit was damaged by either excess input voltage or excess output current, the most likely parts to fail are the transistors, and you can't normally detect that using continuity testing or visual inspection.

If you can't see any broken connections, and you don't have a scope (or wouldn't know what to do with one), you can try replacing all of the transistors, otherwise you'll just need to replace the unit.

>>991398
what software is this?

>>991086
>Binary switches would work well here. It's a switch with multiple positions that mechanically produces a binary output. Pictured here are the thumb wheel variety, there's also rotary ones.
I bought some of them just because I ran across them while shopping for other electronics parts, and hadn't ever worked with them before... I'd just seen them in ancient ~1960's era equipment....

I aint got them yet, will prolly be a couple weeks yet (ordered from china)
I got ten single-digit ones. Do they come apart?--like, if I wanted to use two groups of five?
How does you read them? I doubt the ones Im getting will have any instructions

>>991538
It looks like every circuit, an android app.

>>991576
They come apart and snap together so you can put as many in a row as you like.

Numbered pins are connected to the common based on the wheel position. To connect to an MCU you'd connect common to Vcc and pulldown resistors on each of the numbered pins or vice versa.

>>987270
How do I make a hassler?

>>991651

Building the Widlar Hassler : https://www.youtube.com/watch?v=b0A_5H801rk

Best transistor coming thru, nobleFETs swerve

>>991647
Yea mine arrived today and now I can see how they work...
My confusion was that the little PCB sticking out the back has 10 pads on it...??? five of them are "fake" essentially.
Plus from looking at the PCB you can see how the rotary switch works.

>>991624
thanks m8

>>991367
Learn how a potentiometer works m8

>>991408
Pls, help a anon out, batteries are super bothersome.

>>987082
i got a 230V AC to 12VDC converter. The plan isnt clear about which pin will be 12V and which will be GND. How would you figure it out?

Also i want to switch relais (simple 12V with a coil) and everytime i apply the 12v to the poles it switches like a bazillion times and sounds like a machine gone. What am i missing? Does the A have to be limited with a resistor? Why does it switch on and off then instead of staying on?

>>992529
12V and which will be GND. How would you figure it out?

Multimeter, or something you know the polarity of like LED with resistor.

> i apply the 12v to the poles it switches like a bazillion times and sounds like a machine gone.
Are you sure your converter isn't giving you 12v AC?
Did you wire it up weird with the power to the coil going through the switch?

>>992542
jesus christ, i actually ordered the wrong converter, it gives out AC, i assumed nobody uses 12V AC. Fuck. If you hadn't mentioned it i wouldnt have even checked any more..

>>992561
You can get a rectifier and caps from a bunch of things. It's not entirely a waste.

Trying to find the Y12 admittance parameter as the problem states. Whenever I try it, I end up just getting the negative admittance of the branch with the 1 Ohm resistor in series with the 2/3F and 3/2Ohm resistor in parallel, which is -(2*s+2)/(2*s+5). Since this is obviously not right, can someone give me a hint as to what I should do? Y12 is just I1(s)/V2(s) where V1(s) = 0 (so short) which is, as far as I can tell, simply what I got! I've spent at least 2 hours on this trying it out and can't get it, any help would be greatly appreciated.

>>992686
Literally finishing my power systems class tomorrow. I wish my teacher wasn't shit so I could know what the hell you're doing.

>>992690
This is just a basic lower div network analysis course. I don't know why this problem is giving me such a hard time.

>>992686
I'm getting the same thing as you. Maybe the question is incorrect?

>>992686
Are you quite sure about the definition of Y12?

Because if i2=0 (i.e. no load on v2), then v2/v1 (the transfer function) has that formula with K=1/2.

>>992726
Yes I am fairly sure of it. I originally got what you got but was corrected by someone else. I'll go ask my TA - according to the book, K = -3

>>992744
Well, if V1=0, then clearly I1 has to be -V2 times the admittance of the top branch.

>>992767
Yeah I'm guessing ik not getting something. Will go ask the TA

>>992622
I know rectifiers,
what is a cap though? i'm from germoney.

>>992846
Capacitor.

>>992847
Thanks senpai (familiy)

So /ohm/ I'm a EE student with very little experience in electronic (1st semester) and I'm trying to build a radio with a tin foil variable capacitor I've made. The problem is I can't use 500000000 km of antenna nor I have the skill to make a super-het circuit. And even with a gigantic antenna teh signal will still be piss weak. I want to make an pre-amplifier before the tuning circuit and for the sake of simplicity the plan is to use an op-amp (Or in worst case scenario discrete transistors). Now the question is, what op-amp model should I use? The plan is to hear medium waves (530 kHz to 1700 kHz here) and a normal opamp would shit itself in that frequency.

So i have two GP 12170 (12V 17Ah) batteries that are only putting out around 3v and 4v. Are they dead beyond repair or can i hook them up to a charger (either one for car batteries or a DC supply where i can set a lower Amperage) and bring them back to life again?

If so, will they ever reach their rated capacity again?

Note: I got them out of an USV that has been sitting around for a while, the USV itself refuses to turn on, even if it is hooked to the mains - According to Google it won't turn on if it senses that the battery voltage is too low / missing, so i figured that's where i'd have to start looking.

>>992889
Have you tried a standard radio shack op amp? Stay away from the LM741, but radio shack sells a CMOS op amp (can't remember the model number) that I've had good luck with. I've never used it for frequencies that high but I don't think the frequency range you specified is going to be a problem. I could be wrong, but for the $4 it will cost you to try a regular op amp it's worth a shot.

I've looked on mouser.com and I'm not seeing much for high frequency op amps.

>>988422
Use a heatsink. You'd be amazed how well those things work.

>>992919
Thanks for your reply anon, I'm not american or european so I have no radioshack. That's why I was asking for a model, it's hard for me to find a place to just go there, look for parts and buy them. But I'll try looking for something. Yeah 741 won't work at all (too slow) And I'll be using LM386 for the audio part. I have very very little experience with everything but shit is fun. So many things to do and discover. (Jesus fuck choosing an antenna..) If I don't find a op-amp I think I'll do a simple amplifier like pic. with a 2N3904 which sould work for medium waves. Not having radio shack is really a pain. For you to imagine I'm having trouble finding basic stuff such as ferrite rods and magnet wire (I can only buy in bulk by kilo...)

>>992924
>>992919
>Mouser.com
Never heard of that, I searched for some stupid high slew rate opamps and there are some. I guess I'll ask for help in the lab to see if anyone helps me. The problem is that the assistants work on power applications and don' tknow much about radio/audio. There is one guy who works with pulse width modulation with some big nigger ass transformers but that's is voodoo for me.

>>992933
>>992924
I'm experimenting in high technology antennas. I actually managed to get decent UHF with this shit.

Greetings Ohmians

I've dabbled with electronics in the past, building a radio, those 101-in-1 kits. Most of that has been instructional, i.e. solder this, connect that there etc, but it never explained what was going on under the hood. This is what I want to learn.

I don't really want to get a kit at the moment. My room's already a mess with half finished projects here and there. Also with semiconductors there seems to be so many variants of a single component, e.g. hundreds of diodes and thousands of transistors. I mean to me I thought a diode was pretty simple: it lets current through one way. Why do you need so many? And I get the feeling that even if I get myself a starter kit, I will still need to keep on buying new components and end up with a massive stockpile of unused ones.

I specifically want to learn about analog electronics, obviously the basics at first. I good target for me would be to understand and build a simple amplifier using discrete components.

I've used SPICE before for basic simulation before. And I was wondering if I could use it for learning electronics from the ground up in place of physical components. Could I work through a standard electronics textbook and do the exercise and try out the circuits all through SPICE? Is it accurate enough?

I obviously do intend to build physical circuits eventually, but I just don't want to invest in the outlay for the moment.

how do i get a band saw using AC power to become variable speed? would hooking up a light dimmer to it work?

>>992948
If you're just starting out, use this instead.

http://www.falstad.com/circuit-java/

>>992951
>how do i get a band saw using AC power to become variable speed? would hooking up a light dimmer to it work?
1. how much power is the motor?
2. you need to find out what kind of electric motor it has. specifically, if the motor is a brushed/brushless or synchronous type. most likely synchronous

brushed motors are easy to add SCR throttles onto, you can get the controllers cheap online. Harbor Freight makes a router speed controller for $24 that works great.

for a brushless motor (this is not likely what you have, but anyway) you may be able to modify the control circuit to add variable-speed, or use another controller entirely

synchronous motors are a pain in the butt.
to get them to run at variable speeds, you need to use a rotary phase converter. they cost $$$ to buy but some people do build their own.
an SCR controller will not work on a synchronous motor, at all.

if you MUST have variable-speed and the saw has a synchronous motor right now, the cheapest way is to put a brushed motor on it instead so you can use a cheap SCR controller. The brushes will wear out if you use it a LOT, but for a home/hobby use saw that could take a very long time--just buy a couple sets of motor brushes right away and store them somewhere in case you cannot get them in the future.

>>992561
>jesus christ, i actually ordered the wrong converter, it gives out AC, i assumed nobody uses 12V AC. Fuck. If you hadn't mentioned it i wouldnt have even checked any more..
you *always* gotta read the labels when you buy wall warts, friend
they come in all kinds of voltages (I've seen one that was 72 volts) all kinds of currents (up to an amp or so), both AC and DC, and with different current-smoothing and overload-protection/cutoff features built-in
and they all look the same on the outside

>>992948
>if I could use it for learning electronics from the ground up in place of physical components.

Sure, a simulator can be a great learning aid, but one of the problems with simulators is that if you put garbage in, you'll get garbage out and realizing the results are garbage can be difficult for a newbie.
While the accuracy is usually sufficient for newbie-tier circuits, there are still newbie-relevant gotchas. For example, if you try to simulate a basic astable multivibrator, it might be difficult to get it started either due to perfectly symmetrical design (make it unsymmetrical by hand) or because the simulator specifically tries to power it up in a stable manner (disable the relevant simulation options, add your own manual start-up circuit to it).

>>992889
> The plan is to hear medium waves (530 kHz to 1700 kHz here) and a normal opamp would shit itself in that frequency.
The op-amp is only getting audio-frequency input. You can use practically anything in that role.

>>993032
Yeah, that one is for the volume and amplificating the piss weak demodulated signal so I can use something other than a crystal earset/buzzer for sound. (E.g a speaker). I'm talking about a pre-amp (it amplifies the signal from the antena and send it to the tuner, then it gets demodulated and then it gets heard by us) using a op-amp.

>>993032
Uh, ignore that; I thought you were referring to the one in the schematic.

Are you sure that you actually need an RF amplifier? You only need enough voltage to overcome the diode's drop (the diode should be a germanium type, ~0.1V drop). You don't need current amplification.

The classic class-A single-transistor amplifier will work fine provided that the transistor has enough bandwidth.

>>993040
The current of the signal will be piss weak without an amplfier won't it? That is why I either need to use a buzzer/piezo earset (high impedance, responds to weak signals) or increase the signal. Normal radios have a pre-amp stage (but they usually are super-het radios not "crystal" sets like this). I was trying to avoid using discrete components because I usually make a mess. (Lel my tuning capacitor is made of cardboard, pvc film and tinfoil and I haven't even made the inductor). Don't you know any op-amps fast enough for a radio signal?

>>992951
If it's a universal motor (rotor and stator coils), you just need a chopper circuit (like a lighting dimmer, but it needs to be able to handle inductive loads, and most lighting dimmers can't).

If it's a synchronous or near-synchronous (e.g. squirrel-cage) motor, you can't vary the speed without varying the AC frequency, which requires a variable-frequency drive (VFD).

>>993044
And the amp also would serve to decrese antenna size. Medium wave technically needs a whopping 55m of antenna.

>>993044
> The current of the signal will be piss weak without an amplfier won't it?
Yes. For the RF stage, it doesn't matter. If you buffer the AF, the only load on the RF stage is the demodulator's filter capacitor (and you're interested in its impedance at AF; at RF, you want the input impedance to be much higher than the capacitor's impedance so that the RF is filtered out).

> That is why I either need to use a buzzer/piezo earset (high impedance, responds to weak signals) or increase the signal.
You can boost the output current at the AF stage, which makes life a lot simpler. The only reason for amplifying the RF is if the voltage is too low to forward-bias the diode.

>>992948
> I've used SPICE before for basic simulation before. And I was wondering if I could use it for learning electronics from the ground up in place of physical components. Could I work through a standard electronics textbook and do the exercise and try out the circuits all through SPICE? Is it accurate enough?
Mostly.

Apart from the issue mentioned in >>992969, the lack of parasitics can also cause problems (e.g. applying a near-perfect square wave across a perfect capacitor can result in 1000A spikes). If seemingly simple circuits result in the calculation grinding to a halt, add a few milliohms, picofarads and/or picohenries of parasitics.

Also: SPICE sucks at mutually-coupled inductors producing equal-and-opposite kiloampere transients out of nowhere.

>>993052
Ah, thanks anon. That cleared it out. And if the RF stage is too weak for the diode, it is better for the signal to amplify it a bit using a transistor or to use a diode with a smaller drop? I'll be using a 1/10 antenna. Still in doubt if loop/spiral or one of those big square ones with lots of big turns.

>>993059
A diode with a smaller drop is simpler. Silicon is ~0.6V-0.7V, Schottky is ~0.15-0.2V, germanium is 0.1V if you can still find one.

>>993075
Ok anon. Many thanks. When the radio is finished I'll post it here.

>>993059
>it is better for the signal to amplify it a bit using a transistor or to use a diode with a smaller drop?
Since you have a power supply, there's also the option of biasing the diode.

Modern-day opamps work well on AM band. For example, gain of 10 would nominally require an opamp with gain-bandwidth product of 17MHz. In practice you need more, but 50MHz amplifiers are still cheap and easy to find. Do NOT use excessively fast opamps.
Keep in mind that the LC tank gives you plenty of resonant gain, but only if you don't load it (much). This means that an opamp (with a low impedance output) before the tank must be matched to it properly. It's easier to just put the (operational) amplifier after the tank. Well, or you can use an amplifier which is better suited for driving the tank, like a basic common emitter BJT amp.

Fucking co-ops at work.

have seen some serious shit.

>>993264
Thank you anon. Yes my lab has a signal generator, DC +-12v and -+5v supply too. The only thing that I don't think they have is a capacitance meter which will make my life a bit harder with my ghetto capacitor. I was hoping to use fixed inductor for better quality (instead of making one) but if I can't get a reliable min/max capacitance I'll have to use 2 degrees of freedom in the tuner. I'll test a simple transistor ampifier circuit and some op-amps in the medium wave range using the signal generator and see what it'll be of my life.

>>993487
>>993264

Also anyone has experience in antennas? I think I'll make a 1/10 wave antenna.I have copious ammounts of ethernet cables, I'm usign their internal wires for testing antennas.

I'm trying to calculate the output resistance of a transistor circuit.
I've been taught to remove all independent sources and apply a voltage vx and current ix to calculate total resistance. I'm stuck with what to do next since I'm not entirely sure what state the transistor will be in and also how does having capacitors in the circuit affect the total resistance.

>>993497
Can't say I know much about the topic, but the usual small antennas for MW and LW receivers are a ferrite rod antennas and loop antennas. It is also said that they are more immune to man-made noise than the wire antennas, which is nice at low frequencies.

That said, personally I've gotten much better results with 3...10m long wires + high input impedance preamplifiers than with 0.3...0.6m diameter loop antennas with varying amounts of turns (tuned to resonance with a separate capacitor).

You might want to read this: http://www.lz1aq.signacor.com/docs/fa-eng/Weak_signals-mag_loop_engl.htm

>>993695
If it's common emitter or common base amplifier, the transistor collector is often modeled as a current sink having very high impedance. Thus, the collector resistor sets the impedance level. This assumes that you haven't driven the transistor into saturation.
Capacitors result in frequency dependent impedance, but effect depends on how they are connected and whether you consider the frequencies to be high enough for the capacitor impedances to be negligible.

>>993744
Ok anon. I made the decision to ghett-ify my project more and as I couldn't find ferrite rods I'll make my own. I will make tests with clay, paris plaster and papier mache. I'll buy the stuff tomorrow.

Oh man, this PCB (pic related) is a fucking pain. I've been placing and re-placing shit forever and there's still so much free space...
I'm just gonna route it at this point. Getting a higher density out of this would take another day or more and I'm fed the fuck up.

>>994120
What program is that?

>>994124
Altium Designer 15. I highly recommend pirating it. I can't imagine a better compromise of functionality and usability. It has all the features you need for "regular" (i.e. not super-high speed shit) circuits and you can become proficient in it after designing a few low/medium complexity PCBs. And the documentation is googleable, pretty clearly written and practice oriented.

Plus the 3D mode is pretty neat. You'll have to make your own models for non-standard footprints, but it has a wizard for generating complete footprints including 3D models for like 99% of packages that are used in practice.
Also, the integrated supplier search (usable with the cracked versions - pic related) is extremely handy for generating BOMs. You can assign a supplier product to a part on your schematic and the part number will be added to the part as a parameter.

>>994131
Very noice

>>994133
It is indeed. However I think it's waaaaayy too expensive for what it is - a full license costs like 7 grand plus 3 grand a year for updates and access to the Altium "vault" where you can download a huge amount of pre-made components. The downside to it is that it's extremely slow to browse and batch changing of footprints on these "managed components" is impossible as far as I can tell. I used it this time because my Uni has the licenses and accounts, but my next PCB will be vault-free. You really don't gain anything from the vault if you're designing your PCB alone - when working in teams, having managed components can come in handy of course.

>>994136
No worries. I don't have money for that. I haven't used an oficial version of a program in 10 years. Cad and similars included.

>>994131
Dude you should check out http://www.3dcontentcentral.com/ for cheap & free 3d models

>>994120
Who are you getting to assemble that ;^)

>>994136
>it's waaaaayy too expensive for what it is
They sure have hiked their prices. Funny, considering their sw was originally supposed to be the less expensive alternative to absurdly expensive professional tools.
Well ok, its price is still nowhere near the most expensive packages, but it isn't competing with them featurewise, either.

what kind of contact is this? Is there a name for it?

>>994258
Looks like spade terminal. Another name is Abiko terminal.
There are couple of different sizes of them, so measure the width of the terminal before buying.

>>994261
thanks anon

Hi, I've spent around half an hour looking for a decent comparator for my capacitor charger project, the LM714 is just horrible, when used with inputs 0 and +12 it has an output offset of 2 volts, even with a small load, turning the transistor on all the time, the LM311 works fine, but it's almost obsolete, and I killed the one I had left.

>TL;DR
Any comparator which works with positive DC voltage and doesn't sucks?

>>994435
Fuark it, I'll use a voltage divider to get the 2 volt offset below 0.6.

>>994435
The schematic says LM741, which is an op-amp. I'm not sure the LM714 even exists. The TL714 is a single-supply comparator, but it's 5V.

I'm not sure why you think that the LM311 is obsolete.

If I'm solving values in series voltage drops and if it's parallel current divides right?

Do I just use ohms law to calculate current divide?

>>994519
Yes

>>994521
OK cool thanks

is this enough to create a decent arc if I charge them and short them? (160 milliohms)

>>994514
Sorry, I meant 741.
>I'm not sure why you think that the LM311 is obsolete.
Well, at least the ones I purchased were pretty old, they came in the ceramic dip package form instead of plastic and didn't seem very reliable, I think I killed one because I joined pins 5 and 6, which I read is recommended if not used, but maybe that version wasn't made for that. The 311 also changed states during a broad voltage range, while the 741 I'm using changes states exactly when the inputs cross.

I'm almost done refitting my circuit to work with an 711. This is the control circuit for a capacitor charger, on the left there's a 555 with 50% duty cycle and adjustable frequency with lots of transistors to control the base of a power transistor on another, on the right there's the charge control, when the input reaches a certain voltage the OP amp goes high, hysteresis kicks in so it isn't bouncing back and forth, PNP transistor activates the other NPN transistors so no current can reach the base of the power transistor, pin 4 (555) goes low and everything shuts down except a green LED, indicating it's charged.

I don't know if I'm missing something or if this circuit can be optimized, there probably are a lot of transistors, maybe someone curious enough wants to take a look and give me some feedback, but if not that's ok. I'm mostly worried about noise affecting the OP amp output causing problems and turning the driver on an off randomly. I'm going go sleep, it's 4:30 am.

>>994567
This is a 2600F 2.5 volts, it can throw a lot of sparks, but no arcs, and that's with 0.7 mOhms, so forget about it.
https://www.youtube.com/watch?v=EoWMF3VkI6U

>>994570
> Sorry, I meant 741.
Right. That's an op-amp, and a fairly old one. Op-amps don't make particularly good comparators, as they don't like having their outputs driven to saturation.

> I think I killed one because I joined pins 5 and 6, which I read is recommended if not used, but maybe that version wasn't made for that.
Pins 5 and 6 can be shorted or left unconnected if not used. Neither should be shorted to ground.

> The 311 also changed states during a broad voltage range, while the 741 I'm using changes states exactly when the inputs cross.
You need 10mV of differential voltage for reliable operation. Also, the inputs should be at least 2V below the positive rail. That can be an issue for low-voltage supply, in which case look for a "rail-to-rail" comparator instead.

> This is the control circuit for a capacitor charger
You'd probably be better off with a buck converter. E.g. the NCP3065 is designed for constant-current (e.g. driving LEDs) and has a reference voltage of 235mV. ICs designed for constant voltage tend to use a 1.25V reference which can translate to a fair amount of power dissipated in the sense resistor. Can switch 1.5A directly or you can add an external transistor if you need more.

>>994567
> is this enough to create a decent arc
Nope. Arc length requires voltage, and those are only 2.5V.

If you want more than a millimetre or so, you aren't going to get it from capacitors; you'll need some form of step-up circuit.

Also: it's far from clear that those capacitors will survive being shorted (2.5V/0.16R=40W, 50F*16R=8s, 0.5*50F*2.5V^2=~156J).

>>994614

OK then, will I get a decent discharge out of them?

if they explode my lust for destruction will be satisfied too

>>992948
http://www.amazon.com/Microelectronic-Circuits-Electrical-Computer-Engineering/dp/0199339139

Literally the book we've used for two semesters, and honestly, one of the only books that's near legible I've ever used. It's still a bit fucked at certain parts, but if you have the patience for it, it'd work fine. It takes a while to "get" it, most of the stuff is really simple and dumb but you just have to see the circuit the right way to calculate what's going on each time. If anyone tells you they understand transistor circuits, they're straight bullshit fucking with you.

>>994435
The output offset on the 741 is 300mv at most. You're doing something really wrong if you're getting 2V.

>>994680
741 does not have a rail-to-rail output. Datasheet says that output being 2V below either rail is the typical value at 2k load, the worst case value being 5V. Even the no load output voltage can be 3V below the rails.

>>994681
The rails are not the offset you fuck. It's typical for op-amps or regulators to have a limit of MAXIMUM VOLTAGE output. The offset is literally the deviation from what the expected output should be. If both inputs on the 741 are grounded, the output will be 200-300mv without nulling. If you use a 12V power supply, you will get 9-10V out, that literally has nothing to do with offsets.

Want to dabble with prop making and such, know nothing of electronics or circuit building

Where do start

>>994685
Technical Theater.

>>994683
The mentioned 2V is a result of the limited output voltage range.
>If both inputs on the 741 are grounded, the output will be 200-300mv without nulling
While it might be that much in some cases, it isn't guaranteed in any way. And generally opamp offsets are referred to the inputs.

>>994687
You are wrong,

Look at the datasheet again. The input offset is the deviation from virtual ground. 1-5mA for the 741

http://www.ti.com/lit/ds/symlink/lm741.pdf

I literally have half a dozen 741s and I can show you indeed that all the output offsets are 200-300mV, and the MAXIMUM OUTPUT is 2-3V below VCC.

I can see you're not using a circuit simulator, why not?

>>994691
>anecdotal evidence (supposedly) based on whopping 6 opamps
>b-but muh simulator

You're joking and not really this stupid, right?
I checked simulator's opinion out of curiosity and mine says 3.48V. I can get the output to 305mV (on single supply) if I violate the allowed input voltage range by connecting the both inputs to the positive supply, though.

>>994570
Your schematic is really claustrophobic, bro. Spread it out.

>>994691
>>994696
kek

>>994603
My capacitor charger has to charge capacitors up to 400 volts,
>>994696
>>994691
>>994687
>>994683
>>994681
>>994680
When the output is LOW the voltage is 2 volts, even with no load connected, I've tried this with several 741s, but the minimum output voltage when used as a comparator is still 2 volts. This was the "output offset" I was talking about, the maximum voltage is around 10 to 11 volts, as the circuit is powered by 12 volts.
The 741s I have are all from TI.

Picture related, one input to ground and another one to vcc, giving a low output value of 2 volts.

>>994755
Anyway, sorry if that discussion was my fault, I'll try to finish the board with the 741 today and test it out.
If I'm not pleased I'll buy a couple of new 311s and replace the 741.

>>994763
I realized the PNP transistor is useless, the first transistor at the output of the 555 will invert the signal, so when pin 4 is low the base of the power transistor will be low too I'm going to print it out, my toner is really low, but I'll try... After that I'll do the final resistor tweaking.
I'm worried about noise, since the circuit will be placed below that huge inductor, but anyway, if it works that's great, if not I'll learn what I did wrong, which is fine too.

>>994755
Well, TI says that the "low" output voltage can be 3V even without any load. The internal schematic suggests that LM741 is one of those amplifiers which will fight back if you try to externally pull the output lower, too. That said, 2V is somewhat high. Usually you get closer to the negative rail.
>one input to ground and another one to vcc
This exceeds the allowed input voltage range. You should stay 3V away from both supplies for guaranteed operation.

>>994763
If you happen to have LM358s or LM324s, their input voltage range covers ground and the output is meant to go pretty close to ground as well.

>>994841
>You should stay 3V away from both supplies for guaranteed operation.
That was just a test but I didn't realize that, when charging capacitors the voltage will be divided so the voltage at the input is 2 volts per 100 volts at the caps, that way it'll always operate around 2 to 8 volts, it still seems to work fine around 2 volts.
I have some LM324 but I didn't want to use a 14 pin IC to just use one OP amp, the LM358 seems a good one, I'll add it to the "buy list" along with the TL072
I made the PCB but I lent my soldering iron and I can't test it.

>>994755
I am a tremendous faggot fruit lala gay homo man. The 741 does go down to 2V if you use a ground. I had only done virtual grounding and a negative supply before.

What a shitty op-amp.

My evidence was based on over 100 students in an EE class. Didn't think the 741 was so shit the negative rail at ground would be that bad.

>>994841
Lm358s are dual op-amps, so one of the first interesting circuit is a voltage controlled oscillator, pic related. They are typically low frequency, but could be used to charge/discharge a cap. An H bridge would work for high voltages.

>>994923
https://www.youtube.com/watch?v=o-8D5t6TJYU

>>995050
> What a shitty op-amp.
The 741 was released in 1968 (48 years ago). Modern variants improve upon the original, but they're constrained by the need to not be worse than the original in any respect, even ones you wouldn't think are relevant.

The 741 isn't a single-supply op-amp. Voltages are referenced to "ground" rather than to the negative rail, so there's no requirement for it to be able to produce outputs or tolerate inputs which are close to the negative rail. If you want a single-supply or rail-to-rail op-amp, there are plenty available.

And it's an op-amp, not a comparator. If you want a comparator, use a comparator. An open-collector output will typically be below 0.5V, and they switch a good deal faster than an op-amp that's been driven to saturation.

Could somebody explain me, what will happen if I just rid off of this diode?

>>995154
RL explodes

can anyone good with audio tell me if i've fucked up here?
i know some of the resistors are non-standard.

Where do you find product documentation?
I'm trying to find the spec sheet of this flashlight driver.

>>995212
>first result on google
>http://led22.ru/ledcat/ledsource/amc7135.pdf
fuck that was hard

>>995213
Hmmm, They're set up so you can change the modes that happen when you click the on button by bridging certain connections. Apart from just finding this picture online where would they document things like that, officially?

Hi

Bought some bluetooth earphones. The thing on the left that looks different from the 2 things (middle and right) does not work. It is supposed to be a button. Can this thing be fixed? Whats wrong with it? looks blown or someshit. Not a electronics guy here.. THANKS IF YOU CAN HELP

>>995211

whats this supposed to be? Phono Preamp?

>>995227
nvm.. the bulged button is the one that works and the one on the right is the one thats broken. ANYWAY IVE GIVEN UP, into the bin it goes

Hey guys, i have a question here.

I want to calculate how much 11 lamps waste in money, so for example for 1 lamp, i multiplicate the watts of the lamp, in this case 58W, for the number of hours that it's used, in this case, 8.5 hours, them i divide the result for 1000 and i get kW/day right? But the electricity bill comes in kWh right? How do i do from here?

Cheers,

>>995240
Multiplying the power in kilowatts by usage in hours per day gives you kilowatt-hours (kWh) per day. Multiplying that by the cost per kWh will give you cost per day.

Note that the usage factor is a dimensionless quantity, e.g. 8h/day = 8h/24h = 1/3.

1kWh/day = 1000Wh/24h = 41.666...W. I.e. a 41.666...W lamp which is permanently on will use 1kWh/day.

>>995243
Ok so, the is 58W, so i do 58/1000 to get kilowatts, and so i get 0.058 kW, it's used 8 and a half hours a day so, 0.058kW*8.5= 0.493 kWh, bt shouldn't it be 0.493 kWh per day since i multiplied it for 8.5 that is the number of hours that it works?

Now i don't know how do i convert this to $$$.

>>995247

find the kwh rate from your supplier and multiply by the kwh. around here it's around 4.5 cents per kwh for the first 30 kwh, then goes up to 6 cents, or whatever.

>>995250
Sure, but if the final result is 0.493 kWh, and imagine that the kWh rate of my supplier is as you said 4.5 all the time, i make 0.493 kWh * 4.5 = 2.2185$, is this what it wastes per day, per hour...?

This is for a company so i don't know the kWh rate right know.

THank you

>>995251

the kwh number is for a given time, so the cost is for the time you used in the calculation to get that result: 8 hours in your calculation.

also, the cost it's not 2.21 dollars, it's 2.21 cents.

>>995237
exactly

>>995254
Ok, it's 2.21 cents per day then, but when i calculed the consumption and it gave me 0.493 kWh it means that the lamp consumes 0.493kWh per day because i multiplied it for 8.5 wich is the number of hours that it's turned on in one day, but then multiplying for 4.5 that is the price of the kWh, that price wasn't per day was per hour?

Thanks,

>>995211

looks ok.
- power LED is backwards
- 10K in series with LED is too big, you get less than 1mA which means it'll be nearly invisible
- there's not much advantage in using regulator in this circuit

>>995259

yeah, it's 2.21 cents total cost per 8.5 hours (1 day of operation)

>>995263
But the kWH of the supplier doesn't come per hour or per day? Or it doesn't make a difference, the only thing that matters to multiply by the usage time ir the consumption of the lamp?

Thanks

>>995261
Haha can't believe I missed that led orientation!
I'm using a 10k led in a similar amp atm and get decent brightness, but I'll tweak it around and see how I go.
The regulator is in there so I can use a cheap unregulated 12v linear wall-wart and not have to worry.

>>995265

you're very confused. the kwh rate (4.5 cents in my case) is for one kilowatt for one hour.

so 58 W for 8.5 hours at 4.5 cents = 58/1000 * 8.5 * 4.5 = 2.21 cents

>>995265
> But the kWH of the supplier doesn't come per hour or per day?
kWh is a measure of energy, not power (which is the rate at which energy is consumed). Each kWh costs a certain amount of money.

If a kWh costs 4.5 cents, then using one kWh per day costs 4.5 cents per day.

>>987613
>still riding the eagle

I too want to get off this wild flight.
Also does KiCAD live up to its reputation?
I dabbled a bit, still getting used to it.

>>995271
>>995270
Oh yeah got it, thank you both 10/10

Ok, so I know this is the kind of question that doesn't get answers, but even some pointers in the right direction would be helpful.

I'm a college student living on campus in California, and the 7k+ that I pay for housing doesn't cover laundry, which costs a little over $4 per load. Since the laundry is paid for by a stored value card that we have to re-fill, I'm trying to figure out a way to either clone the card data to re-fill without paying or to do some other replay attack to do laundry without paying.

After using a USB card reader to figure out what kind of smart card it is (gemclub memo), I used one of those conductive pens to break the pads out to areas I could clip alligator clips to, and then hooked those up to an arduino running a logic analyzer sketch in order to try to skim for the write password.

I ended up getting a lot of strange signal noise (channels that weren't hooked up to anything reported data), a really finnicky and unstable trigger that I couldn't time right, and having too little sample space to catch everything. I ended up ordering one of those $10 USB logic analyzers off of eBay to try and get better results with, but I'm not entirely sure what to do with those results once I get them. I read in the documentation for similar attacks that some smart cards use a form of i2c to communicate.i doubt that'll tell me in plaintext what the write password is, but hopefully I'll at least get raw hex data to use.

The data stored on the card is in pure hex. What do I need to do I'm order to find out what values correspond to the money on the card? How then should I go about editing those values? Is it alright to just manually edit the data, or do I need to clone a legitimate card?

If anyone is willing to help, I'd appreciate it a lot. I know this counts as thievery, but with the ridiculous cost of going here in the first place, I'd like to cut corners where I can.

>>995227

those kind of buttons are fixable, i believe.
- cut off, or peel off, the plastic covering
- pull off the yellow spring material, it should now be loose.
- clean the underside of that material, as well as the metal pads on the PCB with an eraser
- put spring leaf back
- hold it in place with scotch tape

it looks like crazy glue got inside the button, so it may require more than just an eraser, like acetone to dissolve the glue, or a rounded blade to scratch it off.

>>995155
why?, not seeing it

>>995261
I looked at that and thought it was to close circuit with the reverse biased diode at the regulator and light up if you connected the supply backwards

>>995450
Inductive load in AC spiking up the voltage

Ohmlets, what would be a good place to look for a regulable DC power supply schematics?

I've been looking in youtube but i've only seen converted ATX, which i don't have excess of, so i'd like to do one from scratch, sounds like a better experience.

>>995490
Start small and build up to get a grip on how things work
Also this https://youtu.be/wI-KYRdmx-E

Redpill me on DIY kits, i always thought they are soldering excercises for gays?

>>995214
They wouldn't. Literally they'd be giving the schematic and device away to someone else.

>>995490
Depends on what you're doing. I'm going to use a linear regulator with a 12-24V supply. Easy enough to do if you're fine using heatsinks. For a current supply, the lower reference voltage the better. lm317 is a classic regulator.

For set voltages, you can do switched-mode supplies. Anything you DIY would only be good for small amperages, like a negative voltage supply for op-amps, or a boost for 3.7-5V in. For large amps, you can get a universal laptop supply, one with a switch for voltage, and you can get 12-18V 5.5A, or 19-24V 4 amp out. For $20 or cheaper. In theory, maybe you could modify one of the "autosensing" or whatever supplies, and manually set the voltage by replacing a resistor with a pot or something.

At high voltages and low currents, ironically the linear regulator is a ton more efficient and easier to build. Higher currents would destroy you unless you paralled them, which should work.

Hi /diy/, can we get Louis Rossmann added to the youtube section on the pastebin? I think it has a neat channel and has lots of experience when it comes to computer repair and setting your own electronics repair business. Anyway, I leave the channel here, take a look if you wish.

https://www.youtube.com/channel/UCl2mFZoRqjw_ELax4Yisf6w

>>995494
Depends on what you're trying to accomplish, they can be great for complicated stuff like oscilloscopes (~20€ for a very basic, no-SMD-soldering one) or other equipment

>>995494
>2016
>being cis male scum

Yeah, well, if you've never soldered before it's a start.

>>995490
> what would be a good place to look for a regulable DC power supply schematics?
Buy (or otherwise acquire) an off-the-shelf fixed-voltage PSU (e.g. ATX, although that's only 12V) then build a variable-voltage DC-DC converter (e.g. MC34063).

Don't DIY the mains-to-DC side. Apart from the safety issues, high-voltage and isolated converters are significantly more complicated than low-voltage non-isolated DC-DC converters.

As for the schematics, you can find one in the data sheet for any switching regulator. If you want more features, you can replace the regulator IC with a microcontroller.

At the simplest level, a hysteretic buck converter is just the basic switch-inductor-diode-capacitor circuit with the switch controlled by a comparator. You just need either enough hysteresis or enough feedback delay that the switch doesn't end up in the linear region. The main drawback of hysteretic converters is that the operating frequency varies, which makes it harder to filter out EMI.

Guys, sorry in advance if this is the wrong thread to post this in, but my son is about to graduate, and he says he wants to be an electrician. I'm thinking about a Fluke multimeter as a graduation gift. Would a 117 be appropriate? Also, would something else electrician related be a better gift? Thanks in advance, and if this is the wrong thread I'll go to the stupid questions one instead.

>>995629
Does he do electronics as a hobby? If so, you might look into getting him an oscilloscope if you're willing to drop a few hundred dollars on him.

Otherwise, that meter would be fine My dad gave me a fluke 115 about five years ago and I still use it to this day.

>>995634
Thanks for the quick response, especially for this board. The o-scope is a good idea, maybe I'll feel him out and save that for if he sticks with it and actually gets a job! (If he doesn't I'll make him give me the meter back... heh heh!)

>>995639
I think the 117 is around $150 and you could probably snag a hantek DSO5202P for about 250. The reason I suggest an o-scope is because if he's done electrical work in classes or on the side then he probably already has a meter unless he was using the school's.

Best of luck, hopefully he stays with it. Always good to have more people messin with electronics. Kind of a dyin' breed these days. Seems like software and programming are much more popular with the kids for some reason.

>>995644
>>995639
To add to this, I'd only get the o-scope of he's a hobbyist too.

If he just wants to do it as a job and doesn't have a good meter then the meter is definitely better. The company he works for is probably gonna have a meter, if not a good one like a fluke. But they probably won't have a scope.

Either way, it'll be a great gift regardless of what you get. I'm sure he'll use it.

>>995646
It's a good suggestion, I appreciate it. He's like every 18 year old, doesn't know what he wants, but either (or both) will carry over very well into "regular life." I just didn't know what meter was acceptable for an average electrician's toolbox. I'm sure for residential the 117 is more than enough, but I wasn't sure about commercial, and I know just enough to be dangerous. Thanks again anon(s).

Ignore next few images.

>>995629
Good gift, i got the same one for my wife's son, he loves it.

>>994248
It looks to me like most things on there are 0402, with maybe a few 0201. Entirely feasible to do by hand, though it would obviously take a few hours.

>>994570
>>994706
This. Pull apart functional blocks and maybe use net stubs for things like rails and ground. To do it in eagle you have to use the name command, but it's worth the extra second to clean stuff up, especially if other people will be seeing it.

>>995450
Looks like a flyback diode, though those are generally schottky diodes. When switched off, the magnetic field within the inductor collapses, inducing a current. The reverse biased diode, or flyback diode, gives this current a path that isn't your switch.

Without flybacks, the switching mechanism, whether it's a transistor, relay, or actual switch, will get ruined pretty quickly. If you're lucky, it will work the first time, but after that things probably won't go too well.

>TL;DR It's a flyback diode and it's good practice to always have one on an inductive load.

please rate and comment my ESR meter project, and see if it has some obvious errors.

>>987082
Im building a light duty CNC, its made out of 3d printed parts, hence the name. http://www.thingiverse.com/thing:724999

It uses 2 steppers in parallel for the X and Y axis.

I have a few questions about the wiring the stepper. You scale it to whatever size you want, mine will probably be 4 foot by 2 foot.

I will have to lengthen the Stepper motor wires, look at pic related and you can see the bottom stepper wire going through the conduit to the other side.

The far stepper motor on the four foot axis would require an extra 4 feet in wire over whatever length it takes to get to the controller.
The other stepper on that axis would only need the length of wire to get to the controller.

Will I need to make the wire lengths identical to make sure the two steppers are working equally? Or is 4 feet just not enough of a difference to matter?

Also since the X and Y work together, would I need to make the 2 foot axis wires the same length as the 4 foot axis?

There could potentially be like a 5 foot difference between the longest wire and the shortest wires between the two axis

Also what kind of gauge wire would be suitable for say a potential 6 or 7 foot run of wire?
Some people are using Cat5 ethernet cable and using RJ45 connectors to make everything tidy and quick to disconnect.

From reading, its only 24AWG. Is that actually enough to be powering and controlling a stepper motor?

>>996100
>wire length enough
Well, so long as you're not expecting to send information faster than about 80% speed of light / 4 feet = 190GHz, roughly.

Cat 5 or 6 can run some ridiculous watts for higher speeds: http://electronics.stackexchange.com/questions/130341/current-over-cat6-ethernet-cable

Up to about 10W should be doable. If you're ignoring voltage constraints, you could probably run 12V. You can always test. If your supply is trustworthy (like 12V 4A) then you can use a multimeter on the other side of a long cable.

How bad are EE courses? I'm starting Comp Eng next semester and I'm a little worried. Is it actually hard to understand or does it just have a tremendous work load?

>>996078
I like you. You're a determined little try-hard! Here's your troll food.

>>996171
Depends. I went to CC for Engineering and Applied Sciences, and nearly all my classes were incredible.

At uni, quality is a fucking joke. Literally my best classes at uni are about the same as my worst classes at CC. Usually one out of four classes is a total fuck-up, either the teacher has to try and salvage his reputation when he tries to fail 1/4-1/2 the class because of his stupid shit, or the curve is that like a 60 is a B.

The probability course I took was a fucking joke (100% math, not EE, not a good class), and the Electromagnetics course was ridiculous (the physicists get two semester for the same material... sigh).

The analog systems courses weren't too bad (amplifiers/transistors/diodes), and I really liked the more programming classes (Embedded Systems and generic C programming with some light matlab). I heard Communications is a bitch though, but that's more signal processing than digital stuff necessarily, so a CE might not take it.

I really like EE because of how little work we do. The MechEs, CEs, etc are always so busy and it gives me a giggle.

The EE labs were ok, the first amplifier section was single-person labs only, so it was a ton of work, for both the simulation labs in the lecture and the labs in the lab class. The second time around it was much easier, partners for the lab labs, and the simulations were "do the sims and fill in the answers" which involved inspection/calculation/concept questions. I doubt many teachers have that kind of write-up for labs.

>>996171
EE is all about getting your fundamentals down (and knowing when to rely on BASED FORMULA SHEETS). Lots of algebra. As you get further along there's more analysis, and when you get to filters you start doing lapace transforms and the like. If you don't fail out by your junior year you'll probably be fine.

>>996203
>>996204
Thanks.

>>996092
thanks for the explanation

>>992436
I thought I knew how they work. Could you give me a hint?

>>988422
Use thinner enameled wire with more turns, more turns will increase the field while a thinner wire will increase the resistance of the coil, making it run less hot.
Many people fail to realize that, they make a coil which basically has less than 10 ohms of resistance and when they try it out everithing gets incredibly hot. Look at electromagnets in relays for example.

>>996357
Could that also be accomplished by simply adding a resistor to the coil?

>>996445
The magnetic field strength is proportional to both the current and the number of turns.

A resistor will reduce the current which will reduce the field strength Adding more turns will increase the resistance and thus reduce the current, but the increased turns means that the total field strength will remain constant.

>>996320
High end on one pin, ground on the other, what you want in the middle. I'm guessing you fucked it up.

I have a tektronix tds360.

What's the point of the floppy drive?

I never really noticed it

>>996544
The usual purpose would be to store sample data and screenshots.
>I never really noticed it
How is this even possible?

>>996545

that sounds really comfy

Another thing I never noticed was the garage door on the back with an x ray warning

>>996547
It's shit on the 3000 series you can get like 2 or 3 on a single floppy and it takes about 10 minutes to do the save. Faster to use the cursor to write down the values directly and draw it by hand Never used one on a crt scope

What are the green-circled parts in this picture? They look like boxes with two rows of colored dots on them, and only two wires coming out, one each on opposite ends.

I am an oldfag and when I was a kid, *everything* used these things in them.

I would guess that they are resistors, but so far I can't find any web pages that indicate what they were.

Even if they are resistors, does this general type of resistor have a name?

>>996578
Yeh, floppies are shit, use GPIB

>>996720
Ye olde Murikin capacitors. Apparently not much used outside the US.
You can try this: http://www.radioremembered.org/capcode.htm

/ohm/ hate to be asked to be spoon fed but I would really appreciate a point in the right direction.

I took a course this summer for biomed engy and basically have no idea what the fuck is going on when it comes to electronics, pic related. I know anatomy quite well and am learning 3D modeling/mech engineering but know nothing of electronics.

Can I learn fast enough to learn pic related? Please help (in lets say, one months time).

>>996736
u torrent this my friend:
http://libgen.. io/ads.php?md5=0CFC6142AAF0CFE38A3A59D249FC6602

Its the best electronics textbook there is.

>>996736
>>996745
You should be good if you just read through the first 3 chapters. Chapter 3 is about FETs specifically but you will need the prerequisite knowledge to understand

>>996736
Any microelectronics books or the art of electronics.

>>996745
>>996755
>>996761
Thank you anons, I sincerely appreciate it.

If you're curious its towards an effort of building a new type of prosthesis that will literally make it worth using. I can't go into design details of how it takes signals from the user, or how intimate the prosthetic becomes, but the cosmesis is unrivaled, the sizing and custom fitting? Out the window. ADL's will be within ranges, and at the same time, it will weigh less than 4kg (which is essential).

Hopefully there will be a working prototype by next year. I'm in charge of, well you guessed it, anatomy of the human body, skin interactions, etc etc. prevention of infection.

>>996767
You could just buy a VEX or two.

>>994120
>>994131

Fix your silk holy fuck.
What is Q1 doing, why is it hanging out in the middle of the board away from whatever it presumably does?
What's that in the top left and why is it hanging off the board if it's part of the power supply?
Pull your tracks tight on the upper right connector.


Altium is such a wankfest, it's overkill even for the majority of PCB designers working in industry. I'm really fucking tired of people pushing Altium, especially toward hobbiests.

>>994567
Use super caps if you want to dump a fuckload of current. They have retardedly low ESR and large capacitance.

If you just want an arc but no current than sure, make yourself a voltage multiplier and spark away.

>>996767
Sounds pretty cool senpai. I hope it works out well.
t. Hopkins grad student

>>996720
>>996729
Murika was not the only one makin these.
Boxy micas with axial leads were common in the USSR too back in the tube days and at the begginings of germanium semicons. Even in Europe, Czechoslovakia`s national company TESLA made these. Of course that was the east block.

>>996720
>Cyberpunk dominoes

I suspect I have a cold solder joint under my BGA south bridge. I don't have the tools and any experience working with BGAs. Any advice?

>>996097
compact it, it's fucking ugly and that would not matter if it would not be that big

What to do with a tube of 24 74 series dual monostable multivibrators

Sup /ohm/, i have a momentary switch and I want to use it for a looper (for Ableton). i also want use a led with it. When I push the switch I want the LED to go on, and when I push it again I want the LED to turn off. Using a normal switch is out of the question. How can I accomplish this /ohm/

>>996745
>>996767
I prefer "Foundations of Analog
and Digital Electronic Circuits" by Agarval and Lang

How can we deal with the rise on surface mount designs? How can we punish surface mountlets for making us look for equivalent real men trough hole devices?

>>997161
what's wrong with SMDs?

>>997191
not him but theyre a pain to solder by hand

>>997231

desu I'd much rather solder SMD than through hole. There's less cleanup and it seems to go faster.

Is there a field guide out that would help with identifying components in any old circuit I might find?

>>997257
Yep, clipping leads is so 1990

Best (free) pcb design software? I'd like to use Eagle but I don't have the money to buy it (and no up to date torrents of it). Or is the free version good enough to use?

>>997394
Free version is more than enough. It's how I made these.

How do i calculate the thickness of my PCB copper connections?
Is there a formula of Ampere to micrometer?

>>997766
You bust out the less known Ohm's law and calculate the resistance you'll get
From there you work out how much current you need and design accordingly

>>997766
You can try this or any of the alternatives: http://www.4pcb.com/trace-width-calculator.html

Basically you have two options: either you keep the trace temperature rise below some suitable value (10C is popular limit) or you keep the resistance / voltage drop below some suitable value.

Hi /g/,

I'm pretty much a complete noob to this complex field, but like with other things in my life, I've chosen to learn by doing: I'm jumping into the deep end.

I'm working on the schematic for a device that has a few components that require different voltages. Not too unusual, as far as I can tell. Most of my components (microcontroller, wifi module, a few other things) would like 3.3V, but there are 2 new devices that need additional considerations: One requires 5V, the other 6V.

At the moment, I have a single simple buck SMPS that drops down 5V input to the 3.3v required for my main components. I'll obviously need to select a new input power voltage for the addition of the new components... I'm thinking 9v, since it's common to find existing 9v power adapters. I've looked into SMPS designs with multiple outputs but it doesn't seem common.. is there a way to get 9v to 6v, 5v, and 3.3v without using multiple separate power supplies? Would it sufficient to simply use a linear regulator or even just a basic voltage divider placed after a single SMPS that drops to 6v?

I'd appreciate any advice in this regard. Thanks!

>>997834
linear regulators will be just fine, depending on what your new components are. a buck regulator is complex and efficiency doesn't matter if you're running off wall power. are you sure your component *needs* 6v?

>>997855
I'm a programmer, so I tend to try and reduce complexity and redundancies, and I also seek to be efficient when I can.

I'd like to reserve it as an option to run my device off of a battery at some point in the future, so while I'm not going to go too far out of my way for it, it'd be a nice bonus.

Sadly, my component does need 6v. It seems unusual... but the datasheet for it specifically indicates that it's required. It's for a heater circuit internal to a component. The correct operation of the component seems contingent upon this circuit being properly powered.

Thank you for your input!

>>997858
in that case i might recommend getting a 12v wall wart, using a buck to produce 6v, and using a linear regulator on the buck output to produce 3v3 and 5v. depending on what the heater is drawing, a linear regulator could end up needing an ugly heat sink for its stage. the wifi and uc on the other hand won't cause one to dissipate more than 250mW. since you're talking about a heater i assume by "battery" you mean a large battery and not a few AAs so the complexity:efficiency ratio there still weighs heavily in favor of a linear supply.

also, since your 6v might not work well at 5v, maybe your 5v component could work at 6v?

>>997834
> is there a way to get 9v to 6v, 5v, and 3.3v without using multiple separate power supplies?
Convert to 6V then drop to 5V and 3.3V using linear regulators.

Or stick with your existing setup and add a boost converter for the 6V.

A forward converter or push-pull converter can generate multiple output voltages from a single oscillator (they use a free-running oscillator rather than one controlled via feedback), but they still need a separate transformer (or at least separate secondaries) and DC side for each voltage, so you'd probably be better off with 3 buck converters.

>>997862

>>997887

>>997862
Yep, not a couple AAs. I would want to design or locate a proper battery pack for the device in this situation.

I am leaning more towards linear regulators in this case now.

I will look into testing my component to see if it will work at 6v instead of 5v in this case. Thank you!

>>997882
This is close to what I'm thinking of.

The possibility of using a boost converter to 6V from 5V is appealing because it might make it easier to power the device off a 'phone charger'... or any other USB power source... as long as my current consumption is low enough. I think it would require USB 3.0 ports on a computer to provide sufficient current (as 2.0 caps out at 500mA, i think?), but disregarding computers as a power source, it would be neat to make it work with the litany of affordable USB wall chargers.

>>997889
My shameful textual focus error, revealed! A requested new line turned into a post faux pas.

Why did I solve the captcha after quoting, but before writing my message? I am stupid.

>>997191
Lots of shit. First of which is hand soldering. Cleaning is super important as it's much easier to experience problems due to leakage. Parts are easy to lose. Shit's annoying as fuck when you're debugging and have to desolder multiple times, it's much easier to fuck up the foil even when you're careful.

The upside is that things are smaller than ever before, more power efficient, and sleek looking. There's nothing quite as sexy as a well designed SMD board.

help bros

how do i use this stereo audio jack in a circuit?

i know the middle one is suppose to be ground but what about the others?
i hooked up an oscilloscope connected to ground pin and rightmost pin and connected my phone music to the thing and got a sine wave with vpp 20 mV and around 100 Mhz

no idea how this thing works and i cant find anything online

>>997958
One of them is the ground, two of them are the left and right channels, the other two might just be to attach it more securely to the PCB or they may connect to a switch which you can use to detect whether anything is plugged in.

Get an appropriate plug and use a continuity tester to detect which pin is which function.

>>997076
no-one?

>>997958
You know what helps? Plugging in a loose headphone plug and seeing which pins connect to which connection on the plug.

Hey guys, i need to know if this circuit will work with a 12V LED strip instead of the lamp.

Thanks in advance

>>998177
It looks like the circuit simply switches a relay, so yes. You can just switch your led strip with the relay instead of the lamp.

>>998202
So i just need to increave the input voltage on the circuit right?

>>998204
increase*

>>996850
Chill out. No point in making the silk look nice when I wasn't done placing yet, plus I won't be using the silk for component names anyway. Q1 wasn't placed in its final position on the picture you quoted.

Also, I really don't care whether Altium is "overkill". It makes designing PCBs a hell of a lot easier than I experienced it with Eagle. I don't know about KiCAD though, it might be good.

>>998204
No, you just need to connect the LED strip (and its PSU) in place of the lamp. The circuit itself (i.e. everything on the coil side of the the relay) doesn't need to change.

>>998245
But the led is 12v, btw do i power de circuit through where now stands the +5V or the 220AC?

>>997076
Flip flop

>>988809
hey I just bought two of these to use with arduinos but apparently they can be used stand-alone? Or maybe I got it wrong?

I just need to make a GET request to a fixed address when an input signal is received, do I have to use an arduino to "drive" it or does it have a programmable chip that can do the same?

>>998254
>hey I just bought two of these to use with arduinos but apparently they can be used stand-alone? Or maybe I got it wrong?
ESP8266 modules do have their own processor and memory on-board.
https://en.wikipedia.org/wiki/ESP8266
processor:
32-bit RISC CPU: Tensilica Xtensa LX106 running at 80 MHz

I am that guy whose post you quoted. I have a few different kinds on hand now (they have been filtering in through the mail) but I still have not had time to play with any of mine yet.

>>998250
The 220AC is where the 12V for your leds will be and the lamp is where the led strip will be.
However the circuit itself needs 5V.
This circuit through the relay "controls another circuit" so to speak. Basically it switches a switch. It doesn't matter what that switch switches.
If the led strip has a good enough power-supply you could simply use a voltage regulator to make 5V from 12V for the circuit. Or you can use a separate psu to power the circuit. You can't apply 12V to the circuit.

>>998252
nice, thanks!

>>998292
Well, he can apply 12 volts to the circuit, the 555 works fine up to 18 volts, he might have to change those 10k resistors by 22 k ones though, but the circuit operation will still be more or less the same.

What does the circuit do btw?

>>998299
>What does the circuit do btw?
From the looks of it the first stage will simply pull the trigger pin on the 555 low for a small amount of time when the sensor gives a high signal.
I don't remember how the 555 operates exactly, but if I would have to guess this circuit simply switches the lamp on for a time when the connected sensor signals it. There's a 10uF cap and a potentiometer for setting the time/charge rate. Looks like a general one-shot 555 timer circuit to me.
Then the last stage simply drives the relay.
>555 works fine up to 18 volts
Good to know. I told him not to connect it to 12V due to the fact that he might need to modify some resistors. You are probably right he should simply try it out with 12V. He doesn't look like someone who knows what he's doing though.

>>987082
/ohm/ I'm making plans to make a robot that I want to have a reasonable distance of at least 15 miles. I've come to the reasoning that Instead of trying to squeeze the life out of traditional battery's. that it is possible to run a 8000watt gas generator, hooked up to a series of VERY LARGE capacitors to act as power storage for the pumps/motors on the robot, its suppose to be fairly big robot I'm trying to make something that acts as a "pack mule" that can carry lots of supplies for search and rescue, and possibly even fire departments.

The reasoning for gas generators over batterys is as simple as fuel density. you can still get more power/range out of a gas based product, then you can with battery's of the same weight and size.

so before I go and spend a year or more of my life studying electronics past what the basics I know, is this a reasonable feat, to use a generator to make electricity and store said electricity in capacitor banks until the machine needs them?

>>998336
Uh huh. Why capacitor instead of a battery? Or why don't you just keep the generator running?

>>998349
The generator is suppose to be running, but the problem with the generator just running making power, is the legs are predicted to have a large power draw, one that the generator can't supply. a few large capacitors would handle such draws while not dickin up the generators power supply, and not causing any power drops across more important sensitive system, such as imaging and path detection.

I built a four stage Cockcroft-Walton voltage multiplier with the input coming from my function generator set to output a 10Vp-p sine wave. The output of the circuit was 80VDC and was confirmed with an oscilloscope.

I tried touching the output 80V with one hand and my power supply ground with another and did not get shocked. The oscilloscope registered a voltage drop between 5V and 12V depending on how well I gripped the output capacitor and the PSU ground.

Why didn't I get shocked? Why didn't I drop all 80V across my body?

Does learning electronics get easier as you go along? I sort of understand the very most basic concepts, but the further I go the harder things get. learning all the different types of semiconductors and being able to understand a circuit by looking at it seem impossibly difficult. Will I slowly be able to get things without thinking about it, because so far progress has been pretty frustrating.

>>998429
I found it got a lot easier for me when I focused less on theory and more on practical stuff. I bought an oscilloscope, function generator, and DC power supply for my home work bench and loads of components and just started building stuff and probing around to see what was happening.

A lot of the theory made more sense when I could see the real life results.

If it's financially viable for you you should invest in some equipment and start playing around. You don't need the latest and greatest. Vintage analog scopes and function generators can be picked up on ebay for as low as $50.

Of course you should keep your books around too, I do reference mine a lot when I'm building.

>>998271
I just purchased an adapter so it can be used on a breadboard with 5V, pic related.

Now I have to figure out how to actaully hook it up to do what I need, sadly I have to rely on existing tutorials because my electronics skills aren't that good, I also plan to drive it with an arduino micro for now for simplicity.

I just need to figure out how to power it up and how to do stuff after a digital signal is received

>>998433
do you know some good beginner books?

>>998444
The Art of Electronics 3rd Edition.

PDF is around online.

>>998299
>>998321
I see, thank you guys, the thing is, i would like to only use 1 PSU. Any ways to do this?

>>998374
If you want 80V, you need to touch both outputs, one of which is at +40V and the other at -40V.

Also, you're less likely to notice DC than AC.

>>998452
Refer to the answers you were given the last time you asked exactly this question.

>>994923
>>994825
>>994755
It seems to work very well despite the 741, I only messed up with the 555, which has a 60% duty cycle instead of around 40 to 50%, I guess I'll try to fix that by using a pair of diodes.

https://www.youtube.com/watch?v=DdIfi3C-AiQ

>>998450
That book is impossible to read for a beginner. It goes over all of voltage current and resistance in like four paragraphs.

What on earth is this connection?

What kind of diode do I need for this circuit? High voltage side will be max 120V, the gate driver outputs floating 15V.
I have these already can I use them? mouser.com/ProductDetail/STMicroelectronics/STPS2150/

>>998543
Then start with Practical Electronics for Inventors. That one is a bit easier. Keep Art of Electronics around though. It is crammed with useful information and eventually it will prove useful to you.

Guys, if i buy a 220V relay, will it work with a 12V circuit?

>>998662
It should. 220V should be a maximum rating if I remember correctly.

>>998663
All right, thanks for helping.

>>998668
You should look and see if you can find a threshold voltage for the relay though. Basically the voltage required for it to switch.

It might be something like 24V.

I'm more used to small signal electronics than relays and relay logic but I've worked with it a bit.

>>998662
> Guys, if i buy a 220V relay, will it work with a 12V circuit?
Coil voltage or switch voltage?

The switch voltage is a maximum, so it will work fine for switching 12V (mains switches are typically rated for 250V AC, 50V DC).

If it's a 220V coil, you aren't going to be able to drive it from 12V.

>>998565
a boost diode silly

>>998565
Those will be fine.

I built an analog audio spectrum analyzer for my car and when it's powered with no signal attached (and with signal) I get a lot of noise on the 60Hz band with some harmonic noise on the 30Hz and 120Hz band.

I'm wondering if this is coming from the alternator? I don't know what the alternator output voltage or frequency is but I'd assume it's 120V 60Hz like mains and it's stepped down and rectified to the 12VDC.

But if the power in is 12VDC how is 60Hz noise appearing in my circuit? Is it magnetically coupled from the alternator? What would be the best way to deal with this?

New thread:

>>998973
>>998973

>>998854
> I don't know what the alternator output voltage or frequency is but I'd assume it's 120V 60Hz like mains and it's stepped down and rectified to the 12VDC.
It isn't. It's AC but it's low-voltage and the voltage and frequency will depend upon the engine speed. If the noise doesn't change with engine speed, it isn't coming from the alternator.

If it's fixed 60Hz, likely candidates are either a mains inverter running off vehicle power or external mains interference.