[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Home]
Board
Settings Home
/diy/ - Do It Yourself



Thread archived.
You cannot reply anymore.



File: 1485251916827.jpg (50 KB, 500x326)
50 KB
50 KB JPG
bump limit reached on old thread >>1267468

https://www.wiki.printf.pl/index.php?title=Pasta (fresh)

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

>What books are there?
Beginner:
Getting Started in Electronics Forrest Mims III
Make: Electronics Charles Platt
How to Diagnose Fix Everything Electronic Michael Jay Greier

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

Advanced:
The Art of Electronics by Paul Horowitz and Winfield Hill

>What YouTube channels are there?
mjlorton
paceworldwide
eevblog
EcProjects
greatscottlab
AfroTechMods
Photonvids
sdgelectronics
TheSignalPathBlog

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

>Where do I get components and lab equipment from?
ladyada.net/library/procure/hobbyist.html
Search the web for "hobbyist electronics sources" to find plenty.
In the US and elsewhere, mouser.com, digikey.com, arrow.com, newark.com are full-line distributors that entertain small orders.

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

>What software should I use to layout boards?
Circuit Wizard
ExpressPCB
EAGLE
KiCad

>Q1. Lithium batteries
Will explode if you abuse them. Read and understand all relevant datasheets and be prepared for catastrophe. See
batteryuniversity.com/learn/
>Q2. My circuit doesn't work. Halp?
Check wiring, soldering, part pinouts, and board artwork if applicable, then post schematic.
>>
File: 1508576046602.png (55 KB, 851x414)
55 KB
55 KB PNG
This thread brought to you in part by the CA3089 FM discriminator.
>>
File: dupont.jpg (126 KB, 1200x1200)
126 KB
126 KB JPG
so I have these dupont wires and I wanted to make some short ones
so I cut out some wire from the center, twisted the ends together and tried to solder
but it wouldn't take
even if I crank it up to 480C these wires wont soak up the solder
what am I doing wrong?
>>
Thread theme

https://youtube.com/watch?v=vtPk5IUbdH0
>>
>>1273132
>not buying a dupont terminal crimping kit
Try adding more flux.
>>
>>1273135
what is the material of the wire?
it's just odd, I never have any trouble with any copper wire that comes in a spool
you can tin that in like 1 second at 350C
>>
Hello /ohm/

I'm looking to create a lissajous pattern generator as my electronics class final project. I acquired a Tek 485 scope from a garbage bin this summer and thought this would be interesting. Overall design I think is going to be two frequency variable sine wave generators, one differentiated to get a cosine. Possibly with some extra stuff to add phase offsets...

Anyways I'm kinda overwhelmed with all of the sine oscillators on the internet. Nothing seems to be quite the right balance of low enough distortion, low enough price, and low enough complexity. Leaning towards a fancy Wien bridge design like pic related.

tl,dr: analog, sine wave, audio frequency, variable frequency, not too high distortion oscillator designs.
>>
>>1273152
Considered a transconductance amplifier? They're kind of awesome.
Also see Barrie Gilbert, “Circuits for the Precise Synthesis of the Sine Function”, IEEE, 1977. http://ieeexplore.ieee.org/document/4240497/authors
>>
>>1273159

OTAs look interesting, but a bit beyond the scope of the class!

And I can't get to that article...
>>
File: sine-cosine.png (4 KB, 529x409)
4 KB
4 KB PNG
>>1273152
This one looks good. More at http://www.massmind.org/images/www/hobby_elec/e_ckt.htm
>>
>>1273164
You might be able to get to it through your university. If not, sci-hub dot cc it.
The Gilbert circuit requires monolithic design anyway, which probably is also beyond the scope of the class. You might be able to get away with matched pairs of transistors, which is either complex or expensive.
>OTAs
It's just an op amp that works in the current domain instead of the voltage domain. Did they say you couldn't use canned waveform generator chips, such as the ICL8038?
>>
File: circuit.png (5 KB, 416x275)
5 KB
5 KB PNG
I don't understand how I can power my motor with voltage, that is too low:
I've got a Vin that changes from +3V to -3V. So I got a NPN for the positive Voltage with +3V on the Collector and for the negative Voltage a PNP with -3V on the Collector pin.
The motor does turn, but not with full speed, although my multimeter says Vout is ~+3V.
I connected the motor to two batteries and it worked fine.

Is my circuit not working? I thought that Vout would have the I of the +3V on the collector side?
>>
>>1273279
You'll lose one diode drop (about 650mV) across the base-emitter junctions, so your output would be more like ±2.35V at small emitter currents, which will increase with emitter current.
Any chance you could put the load onto the collectors and control the transistors separately? Just don't try to turn them both on at once.
>>
>>1273152
You can replace the j-fet contraption in your schematic with a lamp. Certainly bit antiquated, but much simpler and tends to work better. Pic related.
There's another classic producing simultaneous sine and cosine using two opamps, but it has somewhat higher distortion.

And since it's year 2017, microcontroller might be worth considering.
>>
>>1273290
>so your output would be more like ±2.35V
My mistake, I already increased the incoming Voltage, so Vout is ±3V

>put the load onto the collectors and control the transistors separately
I'm not sure what you mean. Are you talking about splitting the signal to positive and negative?
>>
File: halfbridge.png (49 KB, 924x616)
49 KB
49 KB PNG
>>1273301
swap the positions of the transistors, that way you only lose Vce(sat), around 0.2v compared to the ~0.7V Vbe.

Or even better, use MOSFETs instead of bipolar transistors, can get you even less drop.
>>
File: halfbridge.png (10 KB, 573x439)
10 KB
10 KB PNG
>>1273301
Speed and direction, max ±3.8V across the motor. High side BC547B/BD136, low side BC557B/BD135 or equivalent.
>>
File: 1869.png (18 KB, 422x163)
18 KB
18 KB PNG
>>1273291
>>1273152
Good idea, antique lamp is still available.
>microcontroller
You have been memed. No one would do that.
>>
File: 1lissajous.gif (107 KB, 800x796)
107 KB
107 KB GIF
>>1273170
The quadrature oscillators are interesting, but having both sine and cosine from the same oscillator isn't really an advantage for me because I want to have different frequencies.

>>1273176
It's a totally open project, but I think that would be kinda "cheating" in my view.

>>1273291
>>1273340
Yes I've seen these Wien bridges with the lamps everywhere but I thought you couldn't get the right ones anymore (or that it was a pain). The thing that concerns me with all of these is that I need good double pots because there's always more than one resistance that has to be changed in order to change the frequency. Variable frequency is really important to the whole thing.

Are there any circuits that rely on only one resistance to change the frequency? If not I guess I'll go Wien bridge with a lamp and see if that's low enough distortion for that I need.

Thanks for all the suggestions!
>>
>>1273369
One resistance to change frequency? Keep it simple!!! Implement a 555 astable circuit!
>>
>>1273369
>I thought you couldn't get the right ones
The exact type isn't that critical, but of course it is easier to just get something proven to work.

>Are there any circuits that rely on only one resistance to change the frequency?
If you increase the frequency enough, you can use LC oscillators with capacitance diodes and adjust the frequency with one pot. I wouldn't recommend that, though.
The trick ICL8038 & co. uses is to implement a triangle wave oscillator with an adjustable current source (requiring one control voltage) and then reshape the output to resemble a sine. The resulting sine tends to be kinda crappy, though.
For a very limited frequency range, you can also try filtering the higher order harmonics off from a triangle or square wave.
>>
>>1273370
Still need to vary the filter frequency, and the result is pretty spiky.

>>1273374
Who knew making sine waves was such a pain.. Guess I'll order some stuff from mouser and try and make it work with the lamp!
>>
File: Sinegen.gif (3 KB, 300x210)
3 KB
3 KB GIF
>>1273380
>such a pain
Not really. You can get a frequency range of 1:10 with a good stereo potentiometer and two fixed resistors of R/10 in series. After that you have to switch the two capacitors for the next range. The characteristic is f≈1/R of course because all these RC oscillators follow wRC=1 and the required (regulated) gain for the Wien bridge is 3.

If you want a single range, e.g. 20Hz to 20kHz with one capacitor the only reasonable solution is the XR2206 which is current controlled. For Lissajous experiments two independent generators with 200..2000Hz is all you need. A split supply like ±9V or so is nice (no output capacitor) but not mandatory.

Pic: If bulb than discrete..[^.^]
>>
>>1273326
I've tried it out, but I did not get the result I wanted, but thank you!

>>1273337
Your solution worked, thanks a bunch!
I build it with ±3V supplies and I have to adjust the incoming voltage, but I like it so far.
Is there a name for this kind of circuit? I want to read more about it.
I see the name is halfbridge, but I haven't found anything with that keyword.
>>
File: IMG-0841.jpg (890 KB, 1906x2218)
890 KB
890 KB JPG
I am repairing a receiver for an RC car. It is a Spektrum SR3100. It looks like the only issue is this burned out SMD IC.

1. ) I believe it is a thick film resistor?

2.) There are no markings, but I have a working receiver of the same model so I can probably find the Ohms off the other one?
>>
>>1273423
1:10 is certainly sufficient. By stereo potentiometer do you mean a dual gang potentiometer or an audio (logarithmic) potentiometer? Is a 20% dual gang potentiometer sufficient?

200Hz to 2kHz would be nice audio range too, so I make some lissajous music!
>>
>>1273454
1. With all that wire and text "L4" nearby, it's going to be an inductor.
2. The value is going to be difficult to measure.
And finally, inductors won't burn randomly. Something has happened somewhere else first.

>>1273458
He means a dual potentiometer. No idea about the required accuracy; the matching between the sections is more important in any case.
>>
>>1273466
Ya I thought the copper wire seemed out of place for a resistor. The company wont give me any datasheet (its proprietary), but I may by an LCR off ebay for a few bucks.

I will look over the rest of the board to see what triggered the IND to fail.
>>
File: pot.jpg (2 KB, 121x121)
2 KB
2 KB JPG
>>1273458
No, nothing special, a normal linear (not log) double potentiometer. The idea behind the Wien bridge is that both the serial branch and the parallel branch have the same RC time constant which means both Rs must change in the same way. Only then do you get a constant attenuation of 3 which needs a gain of about 3 to sustain the oscillation.
>>
>>1273466
>required accuracy
Not critical, that's what the lamp is for.
>>
File: H-bridge.png (8 KB, 488x276)
8 KB
8 KB PNG
>>1273448
If you search for 'half bridge' you will find a lot, I left off the space. It's one half of a full bridge usually called H-bridge because the circuit looks like the letter H. It runs on a single supply rather than a split (±) supply. The motor can still be reversed but now you need a linear dual (stereo) pot and it must be wired such that while one side goes up the other goes down and vice versa. The picture shows how. The 4 diodes are only for protection of the transistors and can be left off for small supply voltages. 4.5V for a 3V motor should work because you now lose 1.4V compared to 0.7V on the half bridge.

Combining two transistors to increase the current gain is usually called a Darlington configuration and this is a complementary (NPN/PNP) version of it which behaves almost like a single transistor with high current gain.
>>
Hey /Ohm/ I have a battery related question.
How long can a rechargeable battery, say a 18650 cell, be stored before it starts to loose it’s charge?
I’m trying to figure out what I could realistically use to charge then store portable batteries.

Any response is appreciated! (Even memes)
>>
File: 1498987725991.jpg (45 KB, 780x535)
45 KB
45 KB JPG
>>1273499
sixth image hit on duckduckgo for "lithium battery self-discharge"
you're welcome
>>
>>1273454
>Fix the fault which caused it

>Then replace the resistor
>>
File: wien.png (9 KB, 518x288)
9 KB
9 KB PNG
>>1273458
>some lissajous music!
Yup
>>
>>1273553

Looks nice! Any particular reason for the NE5532? Just a good low noise dual op-amp?

Also what exactly is pic related, a 500 $$\Omega$$ trimmer pot? To adjust the gain?
>>
>>1273563
>reason for the NE5532?
Cheap, low noise, can drive low Z loads (even headphones). 500Ω trimmer is for finding the best regulation point for the lamp. I didn't find the U-R data of the 1869 yet. 620Ω output R because 600Ω is the standard for audio generators.
>>
>>1273572
Thanks!

This is the data sheet I found on the 1869:
http://www.mouser.com/ds/2/208/STANDARD%20T-1%203_47-7293.pdf
>>
>>1273580
eu.mouser.com/ProductDetail/VCC/1869
Mouser Part No: 606-CM1869 Minimum: 1000 Stock: 0 Factory Lead Time: 11 Weeks :(
www.elexp.com/ProductDetails.aspx?item_no=0901869 to the rescue..
Found what I was looking for:
http://www.jensign.com/RMAA/Wien_RMAA/
>>
>>1273597
Available on mouser in canuckistan! Gonna order parts and put this together soon, I'm excited.
>>
Anyone have experience in coilguns? What kind of voltage would you recommend for a fairly fast projectile firing coilgun?
>>
>>1273606
Current is what matters, voltage less so. The more voltage you're using the more coil turns you'll need. The best idea is to have a bank of whatever capacitors you choose in parallel, charged up to near their maximum voltage, with some very high current carrying wires between them and in the coil itself. Of course you'll need to mind the maximum current providing ability/the ESR of the capacitors. If you calculate the final kinetic energy of the projectile, then use kinematic equations to figure out the kind of duration projectile will be under the influence of the coil assuming constant acceleration, and use that to figure out the power output and hence the current output, multiply that by a factor of 5 or 10 or so and that will be an estimate for the current you'll need to push. If you calculate the inductance of your coil, you should be able to use the time constant of it (using its ESR and that of the capacitors) to get an idea of whether the coil has too many turns or too few by comparing that time constant to the time your projectile should be under the influence of the magnetic field.

But that's just a theoretical approach of the situation, there's bound to be an anon or two who have experience with coilguns, not to mention a bunch of information on the net. If you really want to, you could try to characterise the entire coilgun in terms of physics, but the magnetic part will be a pain. You'd probably need to take your projectile and measure the force on it from a series of solenoids at a series of distances, which would at least remove the calculation of "magnetic field force vs iron" which has essentially no easy answers. Using a magnet as your projectile makes things a little easier, but still nothing to attempt unless you're well versed in maxwell's equations, the magnetic vector potential, and vector calculus.
>>
File: 20170427175731_1.jpg (243 KB, 1920x1080)
243 KB
243 KB JPG
>>1273606
Maybe you could try using Children of a Dead Earth to get some estimations on the coilgun, it'll do the equations for you. It tells you information about the capacitor voltage and total energy, you'd have to make your own capacitor to those standards because the construction of the capacitors themselves isn't very in-depth. Give me projectile weight, desired velocity, and the material you're using for the coils and projectile and I'll make one for you and post the screenshots. Pic related is an example of what it can do.
>>
>>1273466
>>1273509

The IC doesn't connect to the L4, the two sides connect to the cap? right below it and also to the pin out to servo (so perhaps user error. I got it broken)
>>
I had a diode(?) in my heater pop and kill the low setting and would like to fix it because I use the heater to keep myself warm when I jerk off completely nude in the winter. The part that was being used was marked "6A10 DC" and I would like to replace it with something identical, but I see none with DC on them. What is DC in this case? Is it the manufacturer? I see ones with MIC, or CY, or any other number of letterings, but no DC. I see lots of 6A10 diodes otherwise, just get one of those? I also can't remember which direction it was originally installed.
>>
>>1273714
6A10 is the part number of interest. MIC/CY/etc sound like manufacturing codes that don't change the specification of the part. You could even upgrade it to a 10A10 diode just so that you (hopefully) won't have to replace it ever again.
For this ac application, it probably doesn't matter which way it is installed, since its only purpose is to block one half-cycle of the ac, and either one should be alright.
>>
>>1273643
>Current is what matters
Stopped reading there.
The energy stored in a capacitor increases exponentially with voltage. To deliver more current in a small amount of time and thus creating a strong magnetic field you need more voltage.
>>
>>1273780
Current is what matters with a coilgun's solenoids. You can get capacitors with higher voltages and lower capacitances, or the other way around and the price per energy stored stays somewhat the same. I explicitly stated that the caps should be charged up to near maximum voltage, for the purpose of maximum energy storage. When it comes to building a coilgun, your caps should go in parallel because that lets you get high current, and to some extent, you should prioritise buying low voltage high capacitance capacitors instead of high voltage low capacitance caps. The caps are jellybean parts you can buy at any value, but winding the inductors is something you have to do, preferably not with 500m of wire
>>
>>1273089
Are old smartphone microphones useful?
What can I DIY with them?
>>
File: ECU.png (1.1 MB, 3088x1000)
1.1 MB
1.1 MB PNG
I bought this ECU reprogrammer and the instructions aren't very clear.
It has 4 wires and requires a lamp and a relay. But in not very sure about how to wire the relay to the ECU's pins. Specifically the yellow and blue wires,in the diagram they go before the relay but that can't be right, can it?

Here's some crude instructions I found http://www.ad-by.com/content/images/product/1227259371_Renault%20ECU.pdf

Here's a video of the thing working
https://youtu.be/MBp6rYDpIAk
>>
File: spy mic.jpg (45 KB, 800x800)
45 KB
45 KB JPG
>>1273817
>Are old smartphone microphones useful?

tiny, good quality mics are always useful as you can stick them in tiny cracks in the wall, maybe inside the box for a power socket, and record your sister masturbating to justin bieber.
>>
File: 1509981018578.png (13 KB, 1055x439)
13 KB
13 KB PNG
>>1273492
Thanks again!
I just connected everything and the transistors got quite hot, is it because I connected it like on the left side? Am I supposed to wire it up like on the right side?
>>
File: IMG_20171107_184710_279.jpg (1.63 MB, 2160x3612)
1.63 MB
1.63 MB JPG
>>1273819
So im guessing pin 2 and 18 of the ECU to the back cable on the decoder, then to the battery's negative.
Pin 1 from ECU to the pin 1 on relay and the yellow wire on the decoder.
Pin 48 from the ECU to the pin 2 on relay.
ECU's 37 to green.
And the lamp between 1 on the relay and 47 on the ECU.

Maybe?
>>
File: no.png (8 KB, 142x297)
8 KB
8 KB PNG
>>1273821
This will destroy the transistors.
>>
>>1273823
that is exactly correct apart from every single pin number you listed is different from the diagram.
do you know how to count? 1, 2, 3? no?
the relay is just being used as a shunt or something the ecu probably needs to trick it into thinking its in a car and not on the bench. or maybe the other way around!

just follow the diagram its not rocket science is it?

relay pins look correct but i don't know about automotive relays, you want the two coil pins, either way around shouldn't matter?
>>
>>1273842
physics doesn't apply to this anon, check the other circuit
>>
>>1273859
>>1273842
I've just applied physics, I like it more this way.
Thanks again!
>>
>>1273842
New to electronics, why would this break them?
>>
Hey fellas, I'm a little bit of a brainlet when it comes to this stuff but I'm looking to fix an old IBM Model M2 keyboard. The problem with it is that the capacitors go bad after like two years or something, so you need to replace them with a pair that don't suck. My issue is that I'm on digikey and I'm not sure what I should be looking for. I've got the specs of what I need (one capacitor at 2.2uF 50v, another at 47uF 16v) but looking at the catalog I see that some are for automotive purposes.
Anyway my question is if that really matters at all. Can I just grab two electrolytic capacitors and be done with it? Or is there some difference between what the capacitor is made for/used in?
>>
>>1273921

coz the applied voltages will turn on all 4 transistors at the same time, so current will flow from the top of circuit thru the 4 transistors to the bottom of the circuit, essentially shorting the battery. the current-limiting resistance of the motor is bypassed when it's powered that way.
>>
>>1273924

just buy the cheapest electrolytic. there's no good reason it wont last 20 years since it's not being used in a critical capacity.
>>
>>1273924
I'd also suggest checking out aliexpress, you can get hundreds of capacitors of various values for next to nothing
for example
https://www.aliexpress.com/item/capacitors/32500862928.html
>>
Hey guys I'm getting back into designing a board, but I don't want to use Eagle again. Any good free schematic and board layout software out there?
>>
>>1274008
I've used EasyEDA a couple of times. It's cheap, but the shipping is expensive. The plus side is that you can order as many parts as you want and the shipping is already included.
>>
>>1274008
KiCad is bretty gud, if you won't miss autorouting.
>>
>>1274008
>>1274020
seconding KiCad, it's way easier to use than Eagle and gets a bonus for being free as in freedom
>>
>>1273928
>>1273937
Thanks a ton, I usually just grab cheap sd cards or tools from aliexpress but this is good to know!
>>
>>1274043
Oh, yes, they're very good for cheap components too. Have you had any fake SD cards that stop working after you write a certain amount to them?
>>
File: 1439717042359.png (5 KB, 500x500)
5 KB
5 KB PNG
>>1274044
Ironically I've had more non-chink brand sd cards die on me than chink ones.
>>
>>1274048
Well as long as you're not buying $5 64GB models you should be fine.
>>
>>1273717
Thanks for the reply, going to go ahead and order a replacement.
>>
>>1274053
I recently did just that
there's no risk to it, just test it when you receive it and ask for a refund if it's fake (which it will be)
now you got a free 8gb sd card
>>
>>1273089
Is anyone here insane enough to have tried rebuilding a lithium battery? As in give it new cells, possibly reset the EEPROM on the controller board, and be about their merry way?

>inb4 enjoy your housefire
I'm not ballsy or stupid enough to solder directly to cells. I'm still retaining proper voltage across all cell pairs, but somewhere on the controller it's halting the flow of power.
>>
>>1274107
The battery management controllers are usually off-the-shelf items. Have you checked the datasheets? Some can be reset just by giving them charging-level voltage on the input for a moment. Some may need I2C commands to clear faults. I wager some can't be reset at all without replacing the controller chip.
>>
>>1274107
>controller board
just remove it? none of my lipos have one
I don't see why soldering would even be a problem unless you're an idiot and reverse a connection resulting in a short
>>
>>1274112
Having a difficult time finding any info whatsoever for any component within the battery packs. We're talking laptops between 1994-1998. Though to be honest, the one I have that is 1994 is NiMH, and I haven't even started looking into rebuilding that one yet. I know /diy/ moves slow, so I can come back with some pictures tomorrow if it'l help.

>>1274118
>I don't see why soldering would even be a problem
High heat and lithium-ion is not something I'm willing to try.
>>
>>1274118
The only manufacturer-recommended way of attaching terminals to any batteries is spot-welding. If soldering doesn't cause immediate catastrophic failure, it will degrade the battery's capacity, just the sort of imbalance you don't want in a pack.
You live dangerously, boi.
>>
>>1274123
that's nonsense
you solder to the tab and essentially no heat will reach the lipo itself
unless you literally leave the tip on there for 30x longer than necessary
>>
File: 1494268998796.gif (1.14 MB, 480x358)
1.14 MB
1.14 MB GIF
>>1274126
Then you've got a cold joint, haven't you.
If soldering were an effective and safe way of soldering to cells, even with the low-temperature solder alloys plentifully available, don't you think battery pack manufacturers would be doing that instead of having yet another type of machine in the house?
>>
>>1274126
Did you not read that post? You don't get tabs on raw cells that I've ever researched. So either YOU have to spot weld your own tabs, or try soldering with the great chance of damaging the cell in the best case scenario.

Myself, I'm using conductive copper tape for proof of concept before attempting to fashion up a spot welding setup, if I even get that far.
>>
>tfw measured the output voltage of a slayer exciter with my shitty portable digital scope
>tfw it still works
>>
>>1273898
Here again:

I've build the circuit in >>1273337 (added the 1k resistor) but the motor only turns backwards, when it is attached to the wheel. When I remove the wheel it also turns forward.
Is it because of my signal (used a few Op-Amps in my circuit), or is it possible, that I didn't connect the half bridge properly? Or because the transistor got too hot?
>>
>>1274235
Why didn't you replace the transistors or at least measure some voltages before coming back?
>>
>>1274205

you probably blew out half the protection circuitry and now your scope will kill you if you probe anything over 100V. enjoy your sword of damocles.
>>
>>1274238
>100V
It's one of those $12 AliExpress kits, you're not supposed to put 100V in it anyways. I wound the inductors disproportionately enough that chances are the internal resistance of the scope dropped the output voltage enough to safe levels, especially since the oscillation frequency was barely above audible.
>>
spoiler alert: I'm retarded

So first dumb question, what the fuck are those guarded illuminated buttons called?

>inb4 guarded illuminated buttons

I've tried searching all online stores (australia) and ebay and can only find really cheap shit buttons that don't have those big metal guards.

What search term should I be using?

Also, that 4 colour indicator LED, what the fuck is that called and where can I get one?

TY friends
>>
>>1274243

probably ''electrical recessed plunger pushbutton''
you can probably add ''marine'' to that for precision.
>>
>>1274237
Sorry about that, the negative voltage was at -3.88V and the positive at 3.95V. Also I've replaced the transistor.
So the negative voltage turns the motor, but the positive does not, altough it's higher.
>>
File: expensive shit.jpg (161 KB, 1197x528)
161 KB
161 KB JPG
>>1274243

Google ''$58 for a fucking pushbutton, you gotta be shittin' me"

>>1274248

It's possible the motor self-identifies as uni-directional, and is sick and tired of being told which way to turn by the patriarchy.
What you're describing makes no sense, so you've made a mistake someplace. when it's not doing what it's supposed to, measure voltages everywhere, add them to diagram and maybe we can divine the cause.
include part numbers for the transistors.
>>
File: H_K_Onnes_nl_fr_de_en.jpg (24 KB, 213x452)
24 KB
24 KB JPG
>>1274235
Could be both. Many ways to f things up.
Complete (!) circuit diagram required.
>>
>>1274252
>Google ''$58 for a fucking pushbutton, you gotta be shittin' me"

Ah, yes, welcome to the wold of industrial controls interfaces
>>
File: amp.png (11 KB, 341x341)
11 KB
11 KB PNG
Which is the best/simplest circuit to achieve pic related?

Lets say that I want to amplify a sinus wave in such a way that the output triples everytime the input doubles. Is it possible? I think so, but how?

Do I need an expander or some sort of logarithmic op-amp circuit?
>>
>>1274302
>Do I need an expander or some sort of logarithmic op-amp circuit?

The output would no longer be a sinewave ...

Why do you want to do this? Some analogue synthesizer bullshit?
>>
>>1274303
I have no specific application. I just want to do it and see how it works as a signal amplifier.

Why would the output not be a sinewave?
>>
>>1274243

Allen Bradley illuminated guarded push button switch. Eaton/Cutler-hammer has similar ones.

New or second hand, either way they're fucking expensive.
>>
>>1274312
>Why would the output not be a sinewave?

Because a logarithmic amplifier is a non linear amplifier. a non-linear amplifier distorts the signal, so you get harmonics.
>>
>>1274303
OK, I have thought about it. The output would not be a sinewave, you are absolutely right. It does not matter to me. Is it possible?
>>
>>1274312
>Why would the output not be a sinewave?
Well, you wanted to have this
>output triples everytime the input doubles
shit going on. Let's say your input is a 100mV sine. The output will be +450mV at the sine's positive peak. What should the output value be when the input sine is at its half-value point? Your pic suggests 150mV, which isn't half of the peak's value.

Exponential amplifiers exist, causing this kind of distortion. You can build one around a log amp, but it's not the only option.
If you want the circuit to react to the peak or average value only, then it's called an expander. Like the other anon suggested, they were common in audio stuff.
>>
>>1274302
>output triples everytime the input doubles
what if the input halves or triples?
>>
>>1274331
In my example +/-25mV would make +/-50mV output, while +/-200mV would make +/-1350mV.

This does not make sense as >>1274303
and >>1274317
and >>1274320 have already pointed out.

But I don't know how to explain it any better. I don't know what terms to search for.

Throw all the numbers away: I would very much like to see a circuit that has higher gain on higher inputs. This circuit is not linear and therefore distortion is part of it. But what does it look like?

>>1274320
>isn't half of the peak's value.
Got it... does not work that way.
However, do you have examples for simple expanders? Or exponential amplifiers that work with positive and negative inputs?
>>
>>1274318
Yes, but you need both a logarithmic amplifier and an exponentiator. Because 3^(log2(x)) can not be simplified into either, you need both. It's going to get noisy as fuck.

Easier to do it digital :)
>>
>>1274337
>Or exponential amplifiers that work with positive and negative inputs?

Exponential functions already give outputs for negative inputs, but I'm going to guess that is not what you want. You probably want something which looks like the function x^3 (or 5, 7 etc.). Right?
>>
>>1274348
Probably?! I think so. It makes sense if I think about it.

>>1274338
I am not interested in digital circuits ;)
>>
>>1274351
Take a VCA and a perfect rectifier, use the rectified input for the control input of the VCA.

This gets you somewhat what you want.
>>
File: 311.png (15 KB, 900x406)
15 KB
15 KB PNG
I have a 2 Ohm wire i would like to be heated up. I have a LM311 working as a comparator that should turn the power on and off for the wire with a mosfet, IRF9530.

With a 12V supply the load would draw 6A, but the circuit does not work. This is my main problem. What do i do to fix this?

When i apply a different load that draws 2A it works fine but the mosfet heats up, its should be able to handle 12A without a problem. Only when the load is pulling 1A stays at ambient temp.
>>
>>1273089
I'm not sure if this the right place to post, point me in the right direction if it isn't. I'm entirely new to this sort of thing

I'm thinking about using a CHiP to act as a remote monitoring device. Essentially I want the chip to run a script that I've written whenever a circuit has been either made or broken, i.e. a switch gets opened or closed. From my limited knowledge, I think what I'll need to do is watch for a voltage drop or increase (Does that sound right?) and use that as a trigger for running the script. This seems like a beginner friendly project, am I correct about that?
>>
>>1274462

try moving the load (2 ohms) above the fet.
>>
>>1274486
How would that help? Its a P mosfet anyway.
>>
>>1274487
>Its a P mosfet anyway.

oops. sorry. ignore my post.
>>
>>1274462
What voltage are you getting on the gate of the MOSFET when it's on?
>>
>>1274462
>IRF9530
6^2*.3=10.8Watt

Junction to ambient without heatsink = fuck you, doesn't work.
>>
>>1273783
This is also bullshit though, voltage and numbers of turns increase together as was explained however if you know how electromagnets work the field strength is proportional to not just the current but also the number of turns!

Also there is plenty to be said for trading turns for current in order to keep current heating to a minium, heat is an electromagnets worst enemy.
>>
>>1274489
Hang on a sec while i measure stuff.

>>1274490
Duh, thanks.
Are there any mostfets out there that have lower resistance?
>>
>>1274243
This isn't really a button you buy, you have to buy a kind of collar mounting bit, then the actual button head part you want, maybe a lamp unit bit to make it light up if you want, then the actual electrical switch part all separately, usually you can daisy chain electrical switch unit contactor bits together. Look in the industrial controls section rather than the switches section, look carefully, they don't look like buttons in the catalog and the don't call them buttons either and also sit down before you work out a price.
>>
>>1274497
Yep, you can get a lot lower resistance MOSFETs. Here's one that's 100 times better for 58 cents https://www.digikey.com/product-detail/en/infineon-technologies/IRLR8726TRPBF/IRLR8726TRPBFCT-ND/2354256

N type MOSFETs are generally lower resistance than similar price/size P type.
>>
>>1274482
The little embedded Linux computer? There are two parts to this.
On the hardware side you need to turn that switch closure and opening into a voltage that's 3.3 volts in one state, 0 volts in the other. The easiest way to do this is to ensure that the switch contacts are otherwise unpowered, connect one side of the switch to your choice of input pin, and the other side to ground. Then connect a resistor, about 47k or so (value isn't critical) from the input pin to +3.3V. Now when you press the button, the voltage goes to zero because switch closed resistance << 47k, and when it's high, the resistor pulls it up to 3.3V because switch open resistance >> 47k.
The software side of this can be done any number of different ways according to your choice of language. The quick and dirty way is all in shell scripting by using sysfs. You'll need to find the GPIO pin number for the CHiP pin number, and export the pin you choose to the GPIO subsystem. Then you can set the direction of the pin with something like
echo out > /sys/class/gpio/15/direction
Then you can cat /sys/class/gpio/15/value or whatever and get a 0 or a 1 back. Compare to last value and run appropriate script if different. Add a short sleep to the loop to conserve core and power.
>>
>>1274497
>Are there any mostfets out there that have lower resistance?

Metric fucktonnes, especially if you switch to N-ch. What distributor do you use?
>>
Slapped on a big ass heat sink, it works for now. Interesting.

>>1274499
How about TSM4806? I have some laying around.

>>1274503
I live in the Medieval Central Europe, i have limited options. Farnell is sort of available.

>>1274489
With the 2A load:
GND to G: 0.37V
GD: 11.33V
DS: 0.44V

With the 6A load:
GND to G: 0.34V
GD: 9.71V
DS: 1.6V
>>
>>1274513
>How about TSM4806?
Nevermind that, its an N type
>>
>>1274522
How do i use this?
>>
>>1274514
>Nevermind that, its an N type

So? Does the wire have a fixed connection to ground and are you stuck with +12V?

If you can connect the wire from +12V to drain you can use a N-type. If the 12V power supply is isolated and you can swap it around you can connect the wire from ground to drain (through the MOSFET to -12V).

If you pulse the the MOSFET occasionally you can also use a bootstrap capacitor to get the gate voltage to +24V and use it above the wire.
>>
File: IRF9530.png (86 KB, 993x445)
86 KB
86 KB PNG
>>1274531
Replaced by better version.
Somewhere between 5..10W lost.
Red dot is where you are, depending on temperature.
You need no 100V FET.
A relay needs no heatsink...
>>
>>1274514
Is there a particular reason you can't move the heater wire to the +ve supply instead of ground?
>TSM4806
Those would be excellent as long as you don't blow the gate (Vgs(max) = ±8V, so 5V logic-level), and you can rewire the heater to the high side instead of GND.
>>
>>1274542
>>1274545
Im sort of stuck to low side because theres some extra bit on the wire that is the load that requires the ground.

>>1274543
How do i use this?
>>
>>1274549
Ahhh.
>how do I use thus
You find the line for your Vgs, then you can relate your load's Id to the Vds drop. Summary: that's a pretty inefficient FET for a low-voltage high-current load. You could easily find a better one on Farnell or even at your local electronics components supplier or radio/TV repair shop, if you still have those over there.
>>
Does this look like a good P-fet driver made for 5-24VDC? Also since I assume it's a bad idea to run the micro off a 3.3V linear regulator from up to 24V, any buck converter ICs to recommend me?
>>
>>1274243
I dunno about buttons but for LEDs they're called bezels, and you buy them separately.
>>
>>1274566
How fast do you want to switch? Just 1kHz?

Does it have to be P, or is low side switching acceptable?

Your R11,13 values are pretty darn low.
>>
>>1274566
As for easy buck converters, chips like this are nice, http://aosmd.com/res/data_sheets/AOZ1282CI.pdf, just add schottky, inductor and a few caps and resistors, the MOSFET and control/drive are all built in.
>>
Whats good for an intermediate tools set up for electronics guys?

I have a Antex XS25 25w soldering iron which seems pretty good so far. A cheap helping hand pcb holder with magnifying glass that I'd like to upgrade to something with a bigger lens and I'm going to need a scope as haven't got one.

What do you recommend?

thanks
>>
File: 1508592612000.png (6 KB, 462x563)
6 KB
6 KB PNG
>>1274566
What are your other constraints? Max current, packaging, inductor physical size? If you're just powering the micro and not sourcing any significant I/O current, only sinking, a micropower LDO linear regulator with a low quiescent current wouldn't kill you as far as power waste. LP2950/LP2951 might be good choices here.
The MC34063 is cheap and available, but its max switching frequency is 100kHz, which requires a larger inductor. PDIP available.
XL4005 outputs 5A and runs at a slightly higher 300kHz so slightly smaller inductor, but TO263-5L SMD package and 3-5mA quiescent current.
TPS54232 is good for adjustable output voltages at 2A, 28V max input, runs at 1MHz so you can use a very small inductor. Good quiescent current. SO-8 package.
Most of the bucks in SOT23-size packages are well-behaved, high-frequency, efficient, and easy to design-in. Some are even synchronous, i.e. instead of an external Schottky or FR they have an active switch inside the package, which is extra efficient and convenient.
>P-fet driver
As has happened at least once here on /ohm/, the absolute wrong circuit for some other job is a great fit for the next one. The ill-fated motor driver circuit of >>1273279 is a great, simple gate drive circuit. Here I use it to provide current amplification for a resistive voltage divider. If resistors are more convenient to hand than transistors, you can replace either of the transistors with a fixed resistor, with an increase in turn-on time. This driver won't work well with FETs with low Vgs(on).

>>1274610
18nF gate capacitance sucks
>>
>>1274566
> since I assume it's a bad idea to run the micro off a 3.3V linear regulator from up to 24V
It depends upon the current draw. A linear regulator from 24V to 3.3V will dissipate roughly 6x the power of its load. If the load is a milliwatt, then it doesn't really matter.
>>
File: pp.gif (4 KB, 374x229)
4 KB
4 KB GIF
>>1274566
Both D4 and Q7 will get damn hot at 24V methinks. That is because the circuit has a structural problem: you cannot further pull down the voltage at the base of Q6 because that would kill its base-emitter junction which acts as a z-diode for reverse voltages of more than 7V and the emitter is at about 18V. This leads to the weird values for R11 and R12 which also depend on the actual current gain of Q7. Not good. I would prefer the classic push-pull configuration which has no such problems.
>>
Complete retard here. I want to wire my cabana in my backyard for electricity. What are the basic, general steps to do that?
>>
File: shit.png (19 KB, 444x577)
19 KB
19 KB PNG
super-beginner here, trying to figure this out.

I have a project with a lot of buttons and pots, and i've run into this snag:

I want to be able to read the value of potentiometers and buttons at the exact same time, but if the button is pressed while the arduino reads the value from ADC, the value of the pot drops a bit. I'm assuming this is because the juice required to register a HIGH value in GPio is taken right from what would otherwise be going into ADC.

What can I do to fix this?
>>
>>1274630
Try the QTDDTOT thread. Local electrical codes vary as to how you can do that.

>>1274633
What value is that blue resistor? I bet it's too small. Try something more like 10k.
Better yet, put your switches on the low side, to ground. The micro can then use high-value software-controlled internal pullups instead of that external resistor (set pinmode to INPUT_PULLUP). You just have to remember to invert the value you receive.
>>
>>1274637
hey, you're right. The resistor I was using was about 150ohm, which wasn't nearly enough. a 10k totally did the trick. even a 1k worked well enough, as the pots can tolerate a bit of voltage drop, so even if ten buttons are pressed at once, the pot readings should stay stable. Thanks!

The pullup method is obviously the most ideal, but with the way im scanning buttons I dont think that'd work. Didnt know the atmega had built in resistors though, thanks for that info.
>>
>>1274610
The project is for temperature controlling 24V to a 5V USB soldering iron, because it's smarter than 240V. That gives me 140W or so, so my average duty cycle will end up pretty low, and a high frequency is a good idea. Since it's temperature controlling the frequency doesn't matter, but I'd rather it be above audible so I don't get any whining or humming. The R values are pretty low, R13 gets almost 1W of use on the high cycles, but at the duty cycle it's running that shouldn't be a problem for a 1/4W resistor. Once I have the pieces together (and a FET chosen) I'll increase those resistances to optimise the circuit.

The main reason I've got such an extreme circuit is because of the extremely low load resistance, it means I have to look for FETs with specifically low Rds_on, those that I can find with a value of 10mΩ or less tend to have a gate charge above 100nC. But that's probably being greedy. I'm trying to source parts from Ali so that I don't have to build up a list big enough for free shipping, maybe the IPP100P03P3L-04 from https://www.aliexpress.com/item/IPP100P03P3L-04-3P03L04-MOSFET-P-CH-30V-100A-TO220-integrated-circuit/32830026282.html is good enough, though I've no idea whether this transistor is common enough to trust.

>>1274628
What causes the odd effects with Q7? Is it the base being pulled to ground while the emitter is at 5V or so? I see how your circuit fixes that, which is actually quite clever now that I think about it, but it does charge the FET via the resistor. I'm discharging through one anyway, so I might as well use that, thanks.

>>1274611
That's the sort of thing I was looking for. All I really want is space efficiency, I want a small desktop box with digits on it, not some finned-fanned behemoth.

>>1274625
I plan on using an ATMega328 because I'll be able to transfer the code from an arduino nano clone once I'm done prototyping, since it is damn common and a good starting micro. That's max 12mA, so linear should be fine.
>>
>>1274653
If you're worried about shipping costs, use arrow; they have free overnight shipping above $20, and free ground shipping if you're under that.
>>
>>1274653
>IPP100P03P3L-04
I don't like the threshold voltage on it. You probably do not want low-voltage logic-level devices for this application, however handy they may be for other projects. The emitter voltage drop from our push-pull stage risks not turning it off all the way.
If you can hang with D-PAK, that AOD4184 is good, since it's essentially off when Vgs > -1V, and it's rated for plenty of current at Vgs = -4.5V too.
>free shipping
You might look for sellers on aliexpress that enable combined shipping in your cart. Sometimes you can get a project's bill of parts, by the ten or twenty, and/or in multi-value assortments, for cheaper than you would have with free shipping from a respectable distributor. Just make sure you check the package thoroughly upon receipt.
>>
>>1274513
>I live in the Medieval Central Europe, i have limited options. Farnell is sort of available.
Digikey and Mouser both sell to you. The faggots at Farnell probably want you to use some middleman, but your country is likely to have one (or several). Same with RS.
Then there is a bunch of smaller players, most likely some locals too.
>>
>>1274736
>arrow
Uhh, I already have half an arrow shopping list and it might take some time to complete, through trial and error and all that. Tis the woe of too many project ideas, and their slightly limited selection doesn't help. Getting up to $20 would be tough, unless there's something I should stock up on from them.

>>1274746
I see what you mean, if the threshold voltage gets close to the "c-e forward voltage" of the NPN it could resist too much and cause a voltage drop, which I think I've already accidentally simulated.

>>1274628
Also isn't that FET upside-down? With that circuit (with the correct FET orientation) my simulator gives me quite a bad capacitive effect through the resistor from the 212nC gate charge. But as long as my frequency isn't high enough such that either the FET doesn't get turned on fully at all or that it wastes a significant amount of power compared to the load, I should be fine. The voltages shown are at the MOSFET's gate.
>>
>>1274764
The gate resistor should be pretty smol in that circuit, on the order of a few ohms. Personally, I'd put the resistor on Q9's emitter.
You should try >>1274625 though. It looked pretty sharp at 10kHz in CircuitJS but I didn't run it through SPICE or anything really good.
>>
>>1274775
For the gate resistor to be small, the current through the zener would be way high. Putting the resistor on Q9's emitter is the same as the left-most circuit, assuming the emitter of a PNP is unintuitively the side with the arrow. I'll check out that other circuit now.
>>
>>1274775
*collector
derp
>>1274780
>>
>>1274782
Ah yes, that does work well. I didn't use it earlier when I had an NPN there, but now that I've got a PNP there it stops the voltage at the transistor bases from getting too low, and has a good response time with respect to the gate capacitance. In general, it appears to behave marginally better than that other design: >>1274625, but in both cases there are values to be tweaked.
>>
File: New-Project.png (6 KB, 265x350)
6 KB
6 KB PNG
am i missing something? these NPN transistors conduct in reverse when I put them in this circuit backwards. Both the 2n2222 and 2n4401 light the led regardless of which way I plug them in to this circuit.
>>
>>1274791
Yes, they do that. They just have really poor current gain in the wrong direction, but probably greater than 1. Try again with R1 = 22k and you should observe a difference.
>>
>>1274792
one way gives 6.5mA, the other way around gives 2.5mA
>>
>>1274804
If you're interested, you could measure the base current and calculate the current gains.
>>
>>1274807
Ib = 0.2mA
Ie = 8.5mA

hfe = 42.5?

When I put in a new supply (they're just 3xAAs) from the collector to ground I get different figures.

Ib = 0.2mA
Ie = 45mA
>>
File: IRF7210.png (10 KB, 84x120)
10 KB
10 KB PNG
>>1274628
>5-24VDC
The first victim will be the IRF7210: Drain-Source Voltage Vds -12V
and at 12V the 6V2 Z-diode is nonsense: Gate-to-Source Voltage Vgs ±12V
Moral of the story: if something doesn't make sense, there's probably more.
>>
>>1274748
With farnell i dont have to deal with customs.
>>
File: itcounts.mov.webm (956 KB, 1280x720)
956 KB
956 KB WEBM
I bought some logic gates from ebay. Also had some 555 timers lieing around and now building a counter and then will see where I go from there.
2 bit already working. Now I am just gonna add two more
>>
File: nicebb.jpg (740 KB, 1727x1535)
740 KB
740 KB JPG
>>1275085
pretty good job anon but if you plan on expanding up you need to practice your wire forming technique. spider legs just don't cut it when it gets complicated
>>
>>1275097

and once you have something that complex you need to wire wrap or solder because you will eventually have loose connections that are a pain to track down.
>>
File: moc302x.png (36 KB, 998x447)
36 KB
36 KB PNG
Hello guys. Im working on a dimmer. Ive read up on triacs and optocouplers but i cannot figure out how pic related is going to work. Ive cutted out "not necessary" elements according to description on pic.

So heres how i see it. i turn on the LED inside MOC3021, it lights up and now optotriac inside MOC connects as long as the LED is on. Now how on earth is that supposed to launch the triac outside MOC? I mean theres no voltage betewen gate and the other pin.
>>
>>1275124

the opto element allows current to flow from pins 6 to 4, thus turning on the triac. so it's a switch, not a voltage source.
>>
File: moc302x.png (50 KB, 998x447)
50 KB
50 KB PNG
>>1275135
Okay so basically pic related? im simulating it in LTspice and theres no current flow on the 360 resistor whatsoever. Maybe my models are wrong but they seem to work as intended when i separate MOC and triac and test em that way.
>>
File: fourbit.mov.webm (1.07 MB, 1280x720)
1.07 MB
1.07 MB WEBM
>>1275097
>>1275107
yup thanks.
Now extended it to 4 bits and already had a """bug""" which was just some odd wire. Will clean wiring up a bit more and then move on.
Will get some impressions from ben eater. https://eater.net/
Think I'll build his clock module, too. It's pretty neat for manual or automatic mode toggle thing.
Also got 5 free breadboards for some more modules. Thisgonbegud.webm
>>
>>1275138

well, not exactly: current doesnt loop back on itself. it enters the gate, then goes down toward the load.
>>
>>1275142
yeah true, my bad but the triac outside MOC should now connect nontheless and the voltage on load should be approx 240 V AC assuming the load is much higher than these 2 resistors?
>>
>>1275143
>assuming the load is much higher than these 2 resistors?

what you said is all good except this part. the triac is also a switch, and only needs a tiny bit of current in the gate to turn it on. (when it's on it essentially shorts MT1 to MT2.) so the resistors can be much higher than the load, since they only need to pass a small current.
>>
>>1275147
Hmm yeah. Thanks for the help ill think bout it for a bit longer and then try it out on lower voltages.
>>
>>1275060
Don't order from the US Digikey. Use your local version, like digikey.de.
>>
Where's the best place to get 3:1 heat shrink with various diameters without spending a fortune?

I'd like to have 2mm, 3mm, 6mm, 9mm and a big one somewhere between 15mm and 25mm
I'm hoping some company is selling a collection like this so I could buy one in red black and yellow
>>
File: moc302x.png (45 KB, 998x447)
45 KB
45 KB PNG
>>1275147
I wonder if theres any legit way to calculate the resistance of pic related resistors though. In the datasheet it says that 50mA is enough to trigger the triac. on pic theres basically 830 ohms plus the load. Lets assume that the load is 50W bulb. Im afraid it might burn when i plug it to 240ac.

I know these resistors are going to get shorted by triac turning on but if i turn him on and off rapidly will they withstand the current?
>>
>>1275167

that circuit has been used a million times, and all those times, the resistors have held on. so, they're likely to survive even you.
>>
>>1275169
Hopefully, thanks.
>>
File: Triac_Schematic.gif (8 KB, 474x234)
8 KB
8 KB GIF
>>1275174
Get Application Note AN-780A
>>
File: 1493621256037.png (17 KB, 500x381)
17 KB
17 KB PNG
>>1275167
>turn them off and on rapidly
Triacs latch until the current through the main terminals drops below the specified holding current. If you're trying to "PWM" the input ac half-cycles, you're going to need to find the zero crossing of the input ac and convey that to your control circuitry.
>>
All I've been doing in school is KVL/KCL, calculating terminal resistance of semiconductors, and analyzing single stage transistor amplifiers. I don't really know how to do anything useful.
>>1273089
>The Art of Electronics by Paul Horowitz and Winfield Hill
Would this be a good book to read about some fundamental circuits and their applications?
>>
File: spice.png (26 KB, 829x711)
26 KB
26 KB PNG
When I simulate this in LTSpice, it limits the current through the LEDs to just under 20mA with the 1.33k R1, but when I put it together on a breadboard, I had to lower R1 down to around 35 ohms to get the same current output.

wtf am I doing wrong?
>>
>>1274791
Or you could just put the load on the collector of the NPN, or use a PNP instead. Transistors essentially work by having a voltage and therefore a current from their base to their emitter, so by turning on that transistor, you're putting the same voltage at the emitter as at the base (allowing no current to flow), therefore turning the transistor off somewhat. It works as a current limiter of some kind, and the transistor will never enter saturation.
>>
>>1275214
>All I've been doing in school is KVL/KCL, calculating terminal resistance of semiconductors, and analyzing single stage transistor amplifiers. I don't really know how to do anything useful.
That's plenty useful. You'll get into higher, more applicable levels of circuit structure later on. Once you understand the long-tailed pair, you will be able to understand the op amp, and thus a large and useful variety of linear applications.

>>1275222
Did you match the Vbe and beta of all those transistors? How about Vf of the LEDs? Is your power supply sagging from 13V?
>>
>>1275222
Spice isn't real life, chances are it's the LED forward voltage messing you up. Do a simulation on a single one to get their characteristic curve, and compare that to a few measurements on one of your own LEDs under different resistor values.
>>
This voltage-follower circuit is actually really damn clever. All I have to do is make sure the resistor-divided voltage is above the gate threshold voltage at Vcc=5V and below the gate breakdown voltage at Vcc=24V, and make sure that I've left enough room for the voltage dropped across the BJTs and I'm good to go. It's got ideal switching characteristics for driving a MOSFET since the gate doesn't have to drain or charge through a resistor. Thanks, friend.
>>
>>1275319
As a follow-up to this problem, I'll need a thermocouple to calibrate the iron element/load with, but it seems that I can choose from either a MAX6675-style chip and trust its calibration, or go more the way of making a thermometer circuit with a naked thermocouple and calibrate it (and the ADC) against equations and a few references (boiling and freezing water, maybe molten solders and lead). I'd also need to make a decent thermal junction, probably with some heat-sink compound or thermal rubber sheet, and another type of thermometer to read the reference junction temperature. It doesn't look like the MAX6675 has much of a reference junction, and no temperature reading diode or transistor near the reference junction, so it might not be so good for a mid-high temperature thermometer.

Are the pre-built circuits more accurate than I can expect than anything hand-made will be? Anyone else done this before?
>>
File: 1493760445006.png (39 KB, 408x558)
39 KB
39 KB PNG
>>1275319
You're most welcome, anon.
>>1275395
The monolithic thermocouple amps are probably more accurate than a hand-built circuit would be, especially over temperature. Discrete thermocouple amps require attention to temperature coefficients of various component parameters and tight-tolerance components.
The MAX6675 etc. simulate the reference junction by reading the Vf of an on-chip diode. If you keep the thermocouple connector and the chip the same temperature, you'll be fine. Be careful of copper conducting heat from the power side of your device to the sensing side, etc.
Thermocouple curves don't vary noticeably from sensor to sensor. There isn't much to calibrate if the factory-trimmed ±5degC and 0.5% out-of-the-box accuracy of the device is enough for you. Otherwise, just add an offset to the received value.
>>
>>1275222
>wtf am I doing wrong?
You failed to calculate the values. Also, remove Q2, it is not needed. Horrible diagram style.
>>
>>1275407
In that case that's what I'll get. I'm wanting to get a thermocouple that can take up to 450C and has a tip with low thermal mass, along with some way of tying it to the soldering iron tip with high heat conductivity between them. I thought teflon tape would be good, but it can't handle the high temperatures, and neither can kapton or silicone. I think fibreglass would be too good of an insulator. Maybe some molten thermoplastic? Molten teflon might actually be the way to go, but I would probably need to insulate it in something, like a small borosilicate dish.

>Be careful of copper conducting heat from the power side of your device to the sensing side
I'll probably put it on a seperate board to avoid this, and ensure the thermocouple is anchored close to the chip itself.
>>
>>1275433
You're not going to find a thermoplastic that doesn't degrade at 450degC. Better still would be a fixture of some sort into which the tip is inserted, perhaps on a light spring. That way you can insulate or not you like and still easily maintain contact between tip and probe.
>>
>>1275427
won't removing q2 disconnect the first string of LEDs?
Also I'm well aware my diagrams are shit, working on not making them autistically small and readable instead.
>>
>>1275438
Q7, perhaps? Spread those out a bit so that the labels are less ambiguous and not overlapping, at the very least.
No, all the bases will remain at about the same voltage and therefore receive approximately an equal share of current from your Q1/Q10 current source (?) at the bottom. You only need one string.
That is an odd topology for a current source, btw. What's the idea there?
>>
Hey /ohm/,
Any good reading material recs on a Mach Zehnder modulator? I've to look at its transfer function and relate it to a spectrum it produces when the DC bias I have is minimised, giving some carrier suppression, but I don't know how they're correlated as I've never studied this. i have access to a college library so if there's a book that might've helpful feel free to rec, I am hopefully get my hands on it
>>
File: cc.png (63 KB, 502x406)
63 KB
63 KB PNG
>>1275446
>odd topology
No, it's the simplest way to add some gain. This whole LED string affair was treated before, it's still in the archive. Go to >>1263734 and follow the topic.
>>
File: 1502966604490.png (68 KB, 1317x685)
68 KB
68 KB PNG
>>1275460
Ah, now I get it. Remove the transistors corresponding to Q7/Q9/Q10 on Pic related. You only need one feedback transistor on the outputs, and the rest work in slave mode. You will need to reduce the resistance corresponding to R12 on Pic related as you add more channels, but now it should be more proportional.
You never did say whether you had characterized your real parts, which almost certainly differ from the parts in your vidya and from one another.
>>
>>1275435
The problem with a direct connection is that the thermocouple tip is conductive and is at a nonzero potential with respect to the soldering iron's grounded tip. Though using a small 3V lithium coin cell I could fairly easily power the assembly separately and simply use optocouplers, which now that I think of it is probably by far the best solution. And here I was trying to come up with some sort of DC isolation transformer. I don't think a spring or other assembly is necessary, I should be able to simply press the iron's tip onto the thermocouple while it sits on a piece of FR4.

Most thermocouples (on ali at least) that look to have low heat capacity aren't rated for 450C, like pic related, only up to 350 or so. As far as I know all of those materials are good to at least 700C except for the rubber near the end, will I be fine if I remove only this?
>>
>>1275470
>slave mode
yup
>your vidya
Not mine, I only introduced the current regulation technique. But you are right, you have calculate the values according to load and current gain.
>>
>>1275477
Don't forget the www.google.com/search?q=cold+junction+compensation
>>
>>1275488
I don't have to worry about all that if I use an IC like the MAX6675, which I've already decided to do.
>>
>>1275319
Looking at the IRF9Z24N, it packs 175mΩ and 19nC. There's an IRF9Z24S_L in LTspice with almost twice the on resistance, and it acts fine even at 50kHz, so it looks like a good choice. Anyone used this before?
>>
A very easy question by a stupid dude who's new to this whole thing,
I need to emulate a physical button that grounds a pin digitally.
Essentially:
Button not pressed it's value is Z
When pressed its value is ground
Can I do this somehow with just the digital pins? (send digital 0 (low) as ground, but how do I send Z?)
If I need a transistor, what do I need to look out for that it actuates with the arduino nano's 5/3.3V?
I've never used transistors...
>>
>>1275513
So long as the thing you're connecting to is same or lower voltage than you're running the arduino at you can just use the output pin. pinMode(button, INPUT) and digitalWrite(button, LOW) makes the pin go high Z, pinMode(button, OUTPUT) with digitalWrite(button, LOW) goes low impedance 0v out.

Using a transistor is also possible, for a regular NPN BJT like a 2n2222, connect base to the arduino via a couple kohm resistor, collector to the controlled pin, emitter to ground, then Arduino pin driving 1 gives 0 out, Arduino driving 0 gives Z out.
>>
>>1275513
Did you ask this a while ago?
Tarduino's outputs have direction registers in addition to output state registers. Thus,
- set pin as output, write zero to it = output is pulled to gnd
- set pin as input, optionally write zero to it = output is open/floats

Alternatively, use a small mosfet like 2N7000, connect source to gnd and drain to your button. When you drive the gate high, drain is pulled to ground. When you drive the gate low, the drain floats.
>>
>>1275516
Thanks a lot for that. The Chip I'm running is only runnin goff the arduino's voltage so not factoring in voltage drops within the chip it won't exceed the 5V. Think I'll be well within the ~10% margin of error I have here. The defining factor for this small project is actually to make others interested in electronics and therefore the cost and parts should be held minimal. Thanks for the help.
>>1275517
Kinda but not really. I asked if Digital 0 equals ground back then, but the problem changed as I didn't know how to emulate high Z. Thanks for the help!
>You guys who answer other's questions rock btw
>>
File: spice_comic.jpg (30 KB, 779x294)
30 KB
30 KB JPG
spice gamers are so funny
>>
>>1275525
Not sure I get the joke, unless the joke is that spice circuit dabblers make no effort to select components, and pay no attention to absolute maximum ratings.
>>
I'm fixing an electric coil burner I use for hookah coals. The inline fuse blew, so I'm putting in a new, thicker wire with an inline fuse holder so I can change them when they blow.

The device isn't grounded, and when I was running it without a fuse (which I did for the last couple weeks b/c I'm a lazy piece of shit and wanted hookah coals *now*) it can blow the whole power circuit when the wire inevitably heats up and falls off, grounding to the case. Can I replace the two-prong cord with a 3-prong, and wire the ground to the metal case? Seems that + an inline changeable fuse ripped from automotive shit would be safer.
>>
>>1275601
And also, you know, having not-shit connections, we're soldering or crimping where appropriate now instead of just shitty wrapping wires around.
>>
>>1275601
>an inline changeable fuse ripped from automotive shit would be safer
no
the insulation on your wires will melt before an automotive fuse would blow
>>
>>1275608
You say that, but I just managed to blow a 15amp fuse in this embarrassing monstrosity, take out the breaker, and create a real fireworks show on my counter all at the same time.
>>
File: moc302x.png (8 KB, 754x394)
8 KB
8 KB PNG
>>1275191
yeah i know about zero detection circuit. ive it in plans just not yet. Thanks though.

Now ive one more question. Did i burn my MOC3021? 3.6 / 53.33 = 67.5mA in datasheet it says forward current for input diode is 60 mA. Moreover i didnt start with 53.3 ohms ive started with like 180 and i was gradually decreasing ti to 53.3 but the optotriac still refuses to launch.
>>
>>1275641
Current limit is 60mA, trigger current is 15mA for the MOC3021.
>refuses to launch
How did you measure the triac state?
>>
Laid off bros, what do people do for freelance around here
>>
>>1275702
software development
>>
File: 4chan.jpg (43 KB, 450x450)
43 KB
43 KB JPG
does anyone know how long these jump starters output 1000A's. is it conceivable to get near peak output for an entire 1 second? or at least see a 12v output with >200A for once or 2 whole seconds.

i really wanted to see if i could flash charge ultracapacitors with this. if anyone has a better idea, lay it on me.
>>
>>1273132
>wont soak up the solder
More flux and better quality solder is the answer.
>>
>>1275691
Well i had working circuit with regular triac ( plugged it in to 230ac and the bulb turned on). Basically i made the circuit from moc3021 datasheet with 40W bulb as load. Cannot post pic cus anit at home atm i can do that when i get back.
>>1275188
Is pretty much what i have. Its still working without moc3021 and the gate of triac connected through resistor.
>>
File: test.png (9 KB, 566x346)
9 KB
9 KB PNG
>>1275722
ON-OFF test
>>
>>1275641

your calculation is no good. you take 3.6V then subtract the forward voltage of the LED (which is around 1.5V if it's IR, 1.7V if it's red, 2.2V if its green/yellow) then divide by the resistor. so, for an IR LED, it's 39mA.
>>
File: moc302x.png (8 KB, 565x431)
8 KB
8 KB PNG
>>1275748
Cant try it out at home since i dotn have multimeter atm.
How about pic related method? Stick tongue to A B and check if i feel anything?
>>
File: zapzip.png (29 KB, 265x247)
29 KB
29 KB PNG
>>1275779
Give it a try.
>>
>>1275784
seems to be dead. can feel both of voltages with tongue but when i connect the circuit like on the picture and stick tongue to wires i feel nothing
>>
File: led.png (3 KB, 321x270)
3 KB
3 KB PNG
>>1275786
What about this?
>>
>>1275800
heh. just checked that 1min ago and was about to type. this seem to work
>>
File: 1490002402988.jpg (26 KB, 600x400)
26 KB
26 KB JPG
>>1275779
>since i dotn have multimeter
You are literally working blind, and with mains voltage at that.
>>
>>1275800
>>1275802
but then again isnt burned diode a shorted circuit? or is it open?
>>
File: spicy.jpg (18 KB, 347x356)
18 KB
18 KB JPG
>>1275800
Well the diode works imo. see pic
>>
>>1273089
help me /ohm/

I want to convert my thinkpad x201 to use usb-c charging, quick charge 3.0 and many other standards (even apple iirc) can output 20v ~ 60-80w, my question is, where can i find a board or IC to trick such a power supply into actually providing 20v? I want to try and use a stock power supply so I can still charge my phone off it without needing to manually switch the output
>>
>>1275814
Excellent biometer continuity test, diy at its best. (triggers projections of overprotected children though..)
>>
>>1275525
Hey, spice has a limited selection of components. Since it doesn't model maximum voltage, I can pick any old FET based off its gate charge and on-resistance alone to model what it would be like with a transistor it doesn't model. The IRF7210 had a particularly low on-resistance at 5mΩ, which was good for testing. The 2N3055 was the highest current carrying transistor available, which I was using to test whether turning on the transistors really hard and passing a huge amount of current through them would charge the FET any quicker. But I do see the comedy in the difference in packages.

I guess TO-3 packages are obsolete with respect to TO-220s, since they're inconvenient and I haven't seen a single FET in one while scouring Ali.
>>
>>1275784
Why not put a couple of LEDs (in reverse polarity) in series with a 10kΩ resistor and use that as a voltage indicator? You could even do something with zeners to check what voltage level it's at, but at that point it's a better idea to just buy a multimeter. I bought an AliExpress analog multimeter for $2.50 the other month and it's pretty much useless compared to my $5.00 digital one, but it works. I really just bought it so I could have some redundancy.
>>
Any poor anons out there? I'm looking for electric components in old objects (because right now I can't even afford a meter of copper wire). Any tip for where to search?
pls respond
>>
>>1275882
Sometimes an analog meter is choice especially in high RF environments.
>>
>>1275912

not to mention, they dont give you fake phantom voltage readings when reading AC power. a digital one might show 60Vac, whereas the analog will show, correctly, 0Vac. also, it might show pulses as needle vibrations, when a digital will just jump wildly.
>>
>>1275912
>>1275916
That's true, though I imagine the inductance of the multimeter coil could interfere with some RF circuits. If you had a option to buffer the signal with silicon that might work.

Are there any name-brand analog meters still being made?
>>
This might be a dumb question, but here goes. If I were to make a quick-disconnect thing for a li-po battery with exposed contacts, would I be electrocuted if I were to touch the two contacts? I'm trying to make something like pic related, but with a larger battery.
>>
>>1275925
You won't be. Under 24V DC is generally considered safe. Except perhaps very wet locations. Even then I wouldn't worry about it.
>>
>>1275923
I was talking about taking other DC/AC measurements in environments where there is high RF fields are present around you. Sometimes digital meters loose their marbles.
>>
>>1275916
You just have to think before you read, or bust a cap on it.

>>1275879
TO-3 has the disadvantages of a necessarily live case, complicated/expensive manufacturing, unnecessarily large size, and mediocre thermal performance. TO-220s and D2PAKs are just about as good.
>>
If I slap a big output capacitor (1000u +) on a 12V wall-wart, could anything go wrong?
>>
File: 1480091802028.gif (967 KB, 250x200)
967 KB
967 KB GIF
>>1275929
What if the amperage is kind of high? I might be using batteries with amperage as high as 300 amps.
>>
>>1275938
>could
If unregulated, you might blow the rectifier.
If regulated, you also might blow the regulator.
Odds increase with capacitance and decrease with rated power output of the wall wart.

>>1275940
I still equals E over R. Body resistance doesn't change with available current. The more plausible hazard is a short circuit and all that implies.
>>
>>1275940
>amperage
Hey it's a voltage source, it only draws the current it needs to by the law I = V/R. To 24VDC your skin has 100kΩ of resistance or so, meaning you'll pull less than 1mA of current, give or take an order of magnitude.
>>
File: 1477341026121.jpg (98 KB, 1080x1080)
98 KB
98 KB JPG
>>1275942
>>1275944
Cool, so I probably won't electrocute myself. Now I just need to make the thing work.
>>
File: 1506932977524s.jpg (3 KB, 125x124)
3 KB
3 KB JPG
>>1275814
>>1275800
It works ! wires were fucked since i used super glue and they werent connected properly. Would post webm but its too big. Thanks for everything guys.
>>
>>1275954
>since i used super glue and they werent connected properly
Well on your way to a Darwin Award, good luck!
>>
>>1275704
The limiting factor for trying to pull 1000A out of a pack like that is your cabling, needs to be 12mOhm to get 1000A into a short circuit.

What you want to look for is a cranking amps rating, that's for 30 seconds without dropping below 7.2v
>>
>>1275925
Not a shock risk unless you're putting dozens of cells together. Can be a fire/burn risk, lithiums can put out a pretty huge current if short circuited, so covering the terminals is recommended.
>>
>>1275166
ebay "heat shrink", hong kong sellers will be cheapest. look at the specs and reviews.
or harbour freight or princess auto (automotive / wiring sections)
>>
What are the real advantages of a dedicated soldering station over a cheaper standard stick type soldering iron?
I currently own a Antex xs25 a 25w soldering stick for making small diy electronic synth kits and I’m thinking of upgrading to the HAKKO FX-888D 70W Mains Solder Station. The Antex was £25 and the Hakko is £150 is there really a big difference in quality? Would it be worth investing in the hakko?
I’m also starting to want to test my builds and circuits so I’m looking for a oscilloscope any recommendations for a scope and places to purchase in the U.K.?
>>
>>1276154

£150 workstations are for professionals who work with them 8 hours a day. if you just make an occasional project, then a simple iron will work fine (assuming it has reasonably good tips; doesnt go all black after 10 minutes).
>>
File: dimmer.jpg (3 KB, 251x201)
3 KB
3 KB JPG
>>1276154
Keep your Antex and invest in a good scope (even a used one). The Antex is (more than) hot enough for lead-free but a bit too hot for 60/40 fluxcore type solder wire (flux evaporates too fast). Dimmer helps. I have a 24V version and an adjustable transformer, works like a charm.
>>
>>1276154
The basic idea behind the better (=not shitty) soldering stations is the temperature regulation.
It means that the iron heats quickly to the desired temperature (using the full heater power) and then remains in that temp (using minimal power). If you solder something, the iron stays at the correct temp instead of quickly cooling down.
Simple "x watt" irons use the stated power to heat the tip, meaning that they overheat after a while and that the tip temperature drops quickly when soldering. On the other hand, in order to not overheat immediately, it'll take some time to reach the optimal soldering temp. Overheating also means that they fuck your tips much, much faster than the regulated irons, requiring more maintenance.

Tl;dr. A regulated iron is much more convenient and gives repeatable results easier and with less (or no) maintenance. Whether that's worth the money is your decision.

Some cheaper stations aren't regulated, but basically have a lamp dimmer in series with the iron. They're something between the two.
>>
File: IMG_20171111_160831.jpg (3.32 MB, 3968x2976)
3.32 MB
3.32 MB JPG
Anyone able to help me identify this bulb?
It's from an old 80s era stereo
>>
>>1276154
I picked up one of these last year (same model but without the inspection lamp)

http://www.ebay.ie/itm/SOLDERING-IRON-HOT-AIR-GUN-REWORK-STATION-SOLDER-KIT-730-WATT-W-ACCESSORIES/112410415288?hash=item1a2c2ed0b8:g:T3QAAOSwTUVZsRV5

Thousands of solder joints later and it's as good as ever. I've changed the tip once and replacement tips are dirt cheap
https://www.ebay.ie/itm/Replace-Solder-Screwdriver-Iron-Tip-For-Hakko-936-907-933-Soldering-Station-Tool/282165776691?hash=item41b2641133:m:mM_Z5OD0WdxOgSqfErPz4gw

The hot air is good for any sort of chip rework or even heating heat-shrink with
>>
>>1276176
See if you can find a schematic of the stereo. It will tell you the specs of the bulb.
>>
File: glass-socket.png (47 KB, 236x368)
47 KB
47 KB PNG
>>1276176
Glass socket lamp
T5 W2x4.6d
Voltage and current is printed on the bulb
>>
>>1276206
Cheers.
It says 6/1 on it, I assumed it was voltage / power.
It's hardly 1 amp of current is it?
>>
>>1276159
>£150 workstations are for professionals who work with them 8 hours a day. if you just make an occasional project, then a simple iron will work fine (assuming it has reasonably good tips; doesnt go all black after 10 minutes).
I’m intending to use it quite a lot and hopefully in the future as a job building my own modules

>>1276165
>Keep your Antex and invest in a good scope (even a used one). The Antex is (more than) hot enough for lead-free but a bit too hot for 60/40 fluxcore type solder wire (flux evaporates too fast). Dimmer helps. I have a 24V version and an adjustable transformer, works like a charm
I quite like the Antex seems alright but it does drop temperature sometimes which is quite annoying.
What should I look for in a scope? Do you have any recommendations for makes and models?

>>1276168
>Tl;dr. A regulated iron is much more convenient and gives repeatable results easier and with less (or no) maintenance. Whether that's worth the money is your decision
Sounds ideal,thanks for the detailed explanation. I think i’ll Pick one up tomorrow

>>1276179
This is just the station, no heat gun or magnifier. I picked up a big desk stand magnifier yesterday so I’ve got that covered.

Thanks anons

Oscilloscope next on the list
>>
>>1276206
>voltage / power
Yes, I also have a few that say 12/1
>>
>>1276233
how much money do you have to spend on a scope? if you're poor a used analog scope is your only choice but they work just fine. if you can spend between £3-500 you can get a modern digital entry level scope. the only special feature you're likely to want in that range is a built in logic analyzer. if you don't need that it seems like the ds1054z is the peak of the price-performance curve. $400, so probably £500 after her magesty's tax.

dave jones does scope reviews. i find him annoying but he has a large audience so you can read the comments for affirmation.
>>
>>1276179
I've also been using an 862d+ clone for a couple years
looks like this one with no branding on it
https://www.ebay.com/itm/2in1-SMD-Soldering-Rework-Station-Iron-W-4Nozzle-ESD-Design-LED-Display-862D/371820747844

works great
my only concern is how the wire between the iron and the station is starting to behave, like its getting twisted up internally from repeated use
but that could be from poor handling on my part
>>
>>1275819
Anyone?
>>
>>1276243
Was thinking £250. I have never used a scope before so I’m wanting one for learning and testing/calibrating the modules I make. I have no idea what spec I’d need
>>
>>1276274
if your laptop does not support usb charging you cannot reasonably convert it to do so. unless i'm misunderstanding what you want to do. if you want to nigger rig a 3.0 power bank to your laptop dc jack then it turns out the charging "handshake" is really simple:
http://www.instructables.com/id/QC-30-Trigger/
>>
File: f100_no_lights_on2.png (43 KB, 800x480)
43 KB
43 KB PNG
>>1276279

You can get USB scopes for that that use your computer, but I've read that if you screw up you can damage your computer, which is why I saved up for the Rigol at about $400. If you are on a tight budget can you find a used scope?

Analog scopes are great, but if it's not a storage scope it can be hard to view transients. That's why a digital scope is great because they inherently can store an image and do lots of other cool stuff.

Also, I used a 2-channel scope for years and thought I'd never need more, but now that I have 4 channels I use them quite frequently and love it.
>>
I'm trying to add a 100 watt solar panel to my RV with the possibility of adding another in the future. I bought a 20 amp solar charge controller, but I'm not sure what amp fuse I should get to put between the panel and the controller and the controller and the battery. Would 20 work? Also is anyone familiar with how MC4 connectors work? Can I just assemble them with no other tools besides the assembly tool they sell for them?
>>
>>1276283
>laptop requires 20v and has only analogue charger recognition
>QC3.0 is speced to provide 20v

>cannot reasonably convert it to do so
>>
>>1276355
a 7 coin cell stack also gives you ~20v, try using that to charge your laptop faggot
>>
>>1276284
Thanks. Think I’m going to need another year or so to work out what the hell all the specs mean and relate to what I actually need.
What are the downsides of buying old analog scopes on eBay? There seems to be hundreds of funky vintage military looking ones on there for under £50
>>
>>1276408
if the listing says it works and you can get it with free shipping (or free pick up) then I'd say get one
an old used one will be better than a new cheap piece of shit digital
also craigslist
>>
>>1276408
Main downsides is that they can't take a snapshot of a single pulse/trigger, and you can't get them to tell you exactly the RMS or peak voltage or frequency. But analogue scopes certainly have their charms; they should be better at handling high voltages.
>>
>>1276154
Temperature regulation is kind of important if you're going to do anything much beyond joining wires to stripboard. As a hobbyist or small-time manufacturer, you would be good to go for quite some time with one of the many well-reviewed chinkshit Hakko clones for maybe £40. Annoying Dave also reviews soldering stations from time to time.
>>
Are B0303s-style DC voltage isolators (from ali) any good? I'll be pulling virtually no current through it, no more than a couple of mA.
>>
>Two thirds of the class is failing circuits
>>
>>1276520
That means your shit uni accepted virtually anyone regardless of aptitude.
>>
>>1276525
It means we can't understand the professor or the recitation professors and exams give you 45 minutes to do five circuits using methods that are only (and poorly) explained one lecture before the exam.
You can't even take it before taking calc 2 and other classes that weed out the alcoholic freshmen.
>>
File: circ.png (23 KB, 1248x685)
23 KB
23 KB PNG
Its me again. >>1275167
Ive connected everything with legit wires on test board and it works, sorta. I can turn the light on and off using microcontroller. Heres the issue. When i launch a rectangular signal from microcontroller with 100ms period , 95ms 5V and 5ms 0V the bulb sometimes blinks. its not blinking all the time due to lack of synchronization with 230AC 50Hz but im suprised its blinking at all when theres only 5ms of 0V on the rectangular input.
Any idea or tips how can i fix it so it actually dims instead of blinking?
>>
>>1276529
Profs don't exist to teach you. They exist to facilitate your self-learning. You're not going to get much sympathy here if you're blaming your own failure to study and do exercises on your prof (who a third of the class is fine with).
>>
>>1276536
I know, I've learned more from googling than I have from the class materials.
>>
>>1276536
Classic excuse of the shit teacher
>>
File: relief.png (224 KB, 808x805)
224 KB
224 KB PNG
>>1276550
>Friend goes to office hours
>Professor says "Maybe later" and goes to lunch
>>
>>1276538
i've told this to plenty of college students that never listened but you will get literally almost nothing of value out of all but a few classes. you absolutely have to cultivate electronics as a hobby if you want to be a good engineer.
>>
File: relief 2.png (254 KB, 808x805)
254 KB
254 KB PNG
>>1276551
>Other friend takes Circuits 1 a few years ago
>Gets a masters in math instead after dealing with the professor
>>
>>1276487
Don't use anything from ali if your isolation is safety critical.
There's usually quite a bit of output ripple and noise as well as quite a bit of voltage droop/rise depending on load, if you need a clean supply, use a higher output one + a linear regulator after it.
>>
>>1276553
Everybody who's not already a hobbyist has become a compsci/political science/business major at this point.
I'm getting introduced to useful info like thevenin's theorem, I just have to spend ALL my free time learning the material through means other than the class itself.
>>
>>1276557
A good prof with good lectures still won't substitute self-study. Those lectures only account for 3-4 hours a week, which is much less than the actual time needed to understand and practice circuit analysis.
>>
So I'm working with an AVR644, and the project requires 12 PWM outputs, but only a maximum of 6 at any one time.
Naturally, this is a problem since the 644 only has 6 hardware channels of PWM, but I was wondering: Is there an addressable register that holds the state of OCnx?
Or does OCnx purely refer to the pins?
>>
>>1276555
That makes sense, thanks Mr. 555.

>>1276560
Just use a bunch of shift registers!
>>
>>1276561
How would a shift register help?

At any rate, I would like to change as little as possible, since this is a huge team project, and there would need to be a few small redesigns at this point.
>>
>>1276560
the pwm outputs on atmels are hardwired. i wish they would incorporate those output mapping trees for digital functions that some other ucs have. also if you're using the DIP package you only have 4 outputs broken out.

are the circuits serviced by the pins grouped in any way? if so you can use two 3x1:2 multiplexers (16dip or 16soic) to split your pwms. if this doesn't make sense then give more information on what the pwm signals are doing.
>>
File: Atmega644-pinout.png (308 KB, 961x680)
308 KB
308 KB PNG
>>1276571
I essentially have 6 pairs of motors, and only one motor from each pair needs to run at any time.
I was thinking, if OCnx was addressable, I could have one motor from each pair be on PB4, PB5, PD4, PD5, PD6, and PD7, and then for the other 6 motors I could just read the value from their partner's OCnx register and then write that to their pin.

Although, if I understand you correctly, OCnx only outputs to it's respective pin and cannot be read by the microcontroller internally, yes?
In that case I might have to incorporate a multiplexer then, although my team wants to just implement the PWM in software.
>>
>>1276577
each output compare can generate an interrupt so you can flip different pins in software using those. you really want to avoid having 6 interrupts every millisecond or so though. just use two 4053 muxes ($0.5 ea). alternatively, do you really need to vary the speed independently on all motors?
>>
>>1276579
Thanks!
And yes, all the motors need to be independently throttled.
>>
>>1276580
i assume this is your senior project. if you're not planning on it you should definitely get a pcb made. it's easier than you think and it'll get you a job if you say you're the one that made it.
>>
>>1276583
Close, it's for a college research team.
And yeah, we've already got a few PCB's made, which is why I'm trying to change as little as possible.
>>
>>1276589
do you have a sponsor providing some tight spec enig 16 layer boards for a through hole motor driver project? if so then i understand but boards are so god damn cheap for what you probably need.
>>
File: voltage-divider.png (1 KB, 350x299)
1 KB
1 KB PNG
Okay, /diy/. I need a sort of line level converter to drive a 3 ohm sub off of a 3.5mm stereo on my motherboard. I'll EQ that and mono it using the software controls, but I need to know what value resistors I should use.

Bonus points if someone can show me a way to do it without attentuation, (an NPN thrown in there?).

The part I'm worried a bit about is the sizing of the resistors as I'm planning to not use any other amp besides the mobo's.
>>
>>1276598
Your motherboard won't drive a 3Ω sub, unless you happen to have a computer with a built-in amplifier designed to run low-impedance drivers. The signal a standard 3.5mm device produces is about 1V peak, and they probably won't drive much more than 20mA or so, while plugging your sub in directly will pull a few hundred mA. You'll need to amplify the signal, not attenuate it, though using a 100Ω-20Ω resistor in series might do the trick and get a subwoofer about as quiet as some headphones. What power rating is your driver?
>>
>>1276487
Actually no, they're too expensive. I'll just install 3.3V-5V worth of button-cells instead.
>tfw the optocoupler uses 15 times the current that the isolated IC does
>>
Best intro to electronics/circuits book?

I borrowed make: electronics (charles platt) and enjoyed it probably enough to buy it and the newer one? but wanting to know if there's something newer or better out there
>>
>>1276535
What kind of bulb?

>>1276579
Interrupts are pretty cheap and deterministic on AVR, as long as a little jitter is tolerable and asm is available.

>>1276589
I don't suppose there's room to add a mezzanine board underneath the CPU to do some demuxing? Pic related, the 74LVC1G19, to which you might need to add a couple of hex inverters depending on servo polarity.
Or, if you don't mind a bit of lag, consider a slave AVR with a lot less flash to which you send commands over SPI.
>>
>>1276577
If you have enough time for copying compare result bits from 6 registers to 12 outputs, then you quite likely have enough time to implement software PWM.
>>
>>1276535
TRIACs can only be turned on, they only turn off when crossing through 0V/0A. You'll need to synchronise the AC mains frequency with your micro, the best way is probably with a voltage divider powering an optocoupler, with a rectifier in there for good measure. You'll want to trigger the DIAC at some point into the voltage waveform relative to the next zero-crossing point to determine the dimming factor. But at this point there might be an easier, more digital way around this. You could just use a couple of FETs to trigger the bulb instead of the TRIAC.
>>
I've never really done much /ohm/ stuff before, but I'm redoing the lighting in my game room with LED strips and kind of want to build a really nice little control panel for the dimmer and mount it on my desk. How difficult is a khz range PWM circuit for basically a first project? I remember hearing something about things switching quickly making things weird, is a few khz within the range that does that, or is that only really high frequency stuff, or did I hear wrong?
>>
>>1276698
If it's just one or two channels or colours then I'd use a 555-timer based PWM circuit, they're fairly popular and only a google away. But for RGB or multiple independent lights then I'd run with as many quad comparator ICs as you need, using one as a "triangle wave" generator and the others as the actual PWM comparators for each channel. I recommend this video:
https://youtu.be/QLQrLO0zvDI
to help you grasp how an op-amp/comparator oscillator works. I'm assuming you know how comparators/op-amps, FETs and BJTs work. Also don't forget the comparator's output pull-up resistor if it needs one. Pic related is my take on the comparator oscillator.

A few kHz is no problem provided your components can handle it, you only start getting bad radio-frequency stuff happening once you enter the MHz range, or at least the 100kHz range. Now you might be able to hear the whine of the PWM oscillator if it's in the audible frequency range, which is can be hard to get above, but for a games room it shouldn't be nearly loud enough to be noticed.

Good luck!
>>
>>1276695
I suspect anon only thinks he has that sort of time, and that one half of his robot's going to have Parkinson's.
>>
I am trying to stream a 16x16 video to my arduino led matrix using python but this is the result. is this a problem with my arduino not being able to read the data fast enough or is it a problem on the computer side
>>
File: 123123.png (57 KB, 271x374)
57 KB
57 KB PNG
Anon can you help me? Can't find datasheet for this. N-channel transistor+ 2 resistors. Any analog?
>>
>>1276756
It's a pre biased NPN transistor package, can't find any results for the part numbers there. You can probably guess an equivalent if you've got the schematic of what goes around it.
>>
>>1276535

100ms period is too slow. make it 10ms or less to get the proper effect.
>>
>>1276358
sorry, let me rephrase that, usb-c can deliver 20v, 87w, and is literally designed for laptops. hetro loser
>>
>>1276749
Arduino's serial port has no hardware handshake, so unless you have implemented some kind of software handshake, your PC has no way to know whether your Arduino SW is having problems or not.
That said, the USB to UART bridge sends stuff at some fixed speed and at some point stops accepting more data until the old stuff is sent.
Either way, the problem is in your PC software.
>>
How would one make a circuit which utilizes a servo that turns 180 and returns back to the starting position? In my case there´s a small cup attached on the servo that catches; rotates 180; drops the goods; returns.

Obviously this could be done easily with an arduino but I want to build it piece by piece (Or should I just use the arduino) There were some no programming servo setups with 555 and a pot but It should be somewhat automatic.
>>
>>1276870
You could definately make it without arduino but the arduino makes it a lot easyer, allthough I get the appeal of doing it without arduino for the challenge, it will probably also teach you a lot about electronics. You could probably just make some voltage controlled 555 servo controllers and translate capacitor charge into servo motion, in that case w discharging the cap would mean turning the servo
>>
>>1276829
i think we're still misunderstanding each other. if your laptop does not support charging from usb out of the box, you cannot get it to charge through a usb port. you can make an adapter cable from a usb port to your dc jack though. if you want to do that my link seems to show an easy way to manage the voltage configuration with as little as a 556, assuming the QC 3 handshakes are consistent across vendors (didn't care enough to check beyond that ON part).

also be damn sure your usb cable and power source can actually carry 4 amps.
>>
File: file.png (46 KB, 257x336)
46 KB
46 KB PNG
Can anyone help me understand the PMOS hybrid pi model? The one at https://en.wikipedia.org/wiki/Hybrid-pi_model#MOSFET_parameters doesn't differentiate between P and N MOSFETS. When I follow it, I get a negative transconductance, which winds up being a current source in the wrong direction, since Vgs is also negative.
>>
File: schematic.jpg (53 KB, 766x926)
53 KB
53 KB JPG
>>1276874
Decided Im just going to use dc motor with endstops and npn switches. Im not that good at designing but I hope you get the idea from the pic. Dc motor swapping direction while using the same power source. Im prettty sure Im way off with my design so please. Help me fix it. Startsignal means just the first 180 degree rotation and stopsignal is 5v when the motor is supposed to spin back.
>>
>>1276900
that circuit looks quite messy but I am pretty sure you just want an H-bridge: https://turnelltech.files.wordpress.com/2013/01/h-bridge.jpg. be carefull with this circuit because triggering both inputs at the same time will short the circuit. also make sure your inputs are pulled low with a resistor when you are not using them. Also when you simulate this circuit keep in mind that a motor's resistance is a lot less than 1K ohms
>>
>>1276687
no idea all thats written on it is 40W 240V E27
>>1276697
yeah i know i need to synchronize arduino with 230AC ill start working on it tomorrow. For now it started to work when i changethe period of rectangle to 10ms.
What i am suprised about is that the bulb is that the bulb is dimming in a weird way. The power delivered to the bulb seems to resemble sawtooth signal with approx 4 seconds period. Could it be lack of synchronization? I thought the phase would be random when i plug it to the 230AC and stay certain, random level of dim untill i unplug and plug in again?
>>1276795
yeah that was it. it works now.
>>
>>1276902
Thats exactly what I want. Thanks!
>>
>>1276905
Ive figured why the power delivered to the bulb resembles sawtooth. Its because the frequency of 230 AC in the socket isnt exactly 50hz and when i made some graphs of voltage at bulb with freq close to 50hz i can see that the sine wave changes its shape depending on time.
>>
>>1276887
Pls respond
>>
File: file.png (90 KB, 1598x1199)
90 KB
90 KB PNG
Anyone know why my drain isn't even close to equal to my source current?
>>
>>1276997
it would help if we could see the circuit
>>
>>1277071
relax chang. you have to retool your factory to make shovels for the PLA tomorrow and you need to save your energy. you don't want to be one of the unlucky ones when that gantry gives way again!
>>
I'm doing source transformation problems, and in this example I don't understand why only two of the three current sources are transformed into voltage sources while the 3 amp source in the middle is left alone. Any help is appreciated.
>>
>>1277175
I think I might understand it after reading my notes again. Is this so the circuit is two parts with two resistors and two sources?
>>
>>1277177
Do you get the same answer uf you do it both ways? Could just be the author acting autistic or something
>>
>>1277178
I just tried it and I do.
I guess it's just personal preference.
>>
>>1276916
Using a crystal oscillator on the arduino? if not, the error will be more from your arduino not being accurate to 100hz rather than the line voltage not being 50. For accurate diming do zero xross or peak detection of the incoming AC, then do your dimming timed off that.
>>
>>1277235
The frequency from the wall socket isn't going to as precise as the internal oscillator's, let alone an external crystal. I agree that using zero-crossing or peak detectors is the way to go.
>>
File: 1507978566203.png (54 KB, 1387x447)
54 KB
54 KB PNG
>>1277258
>The frequency from the wall socket isn't going to as precise as the internal oscillator
Internal oscillators are generally only accurate to 1% or so, without regard to temperature, unless recalibrated by the user. 500ppm crystals are accurate to 0.05%. Over the long term, mains frequency is regulated to precisely the nominal frequency over periods of several days, with deviation determined and corrections applied daily. Pic related illustrates measurements vs. standards derived from atomic clocks, conducted in the Netherlands. Corrections are clearly visible on days 17-19 and 41-43.
Granted, it still drifts from ideal over periods of minutes from time to time, and I agree that synchronizing to the actual phenomenon of immeidate interest via a zero-crossing detector is the way to go.
>>
File: motor.png (6 KB, 696x320)
6 KB
6 KB PNG
I am using a LM358 (powered with +-6V) and connected a motor to the output.
Vout is 2.9V, also tried with +4.5V as Vin (resulted in 4.4Vout) but the motor is not moving.
The motor moves, when it is connected directly to the batteries.
Is the LM358 restricting the motor or something?
>>
>>1277339
Yes. Use a half-bridge, see
>>1276902
>>
>>1277342
Wouldn't it also be possible to get a Op-Amp that does not restrict the motor? If so: what do I have to look out for?
>>
File: sI9uM.gif (5 KB, 549x304)
5 KB
5 KB GIF
>>1277339
>>1277345
A damn high current output.Op-amps usually can't push more than a few dozen mA, any motor will probably require hundreds of mA. They're designed for the purpose of transforming analogue signals without unwanted distortion, not driving brute on/off loads. You need to use raw transistors of some kind, probably in a full-bridge.

>>1277342
That's a full-bridge, mate. Also I'd use one with PNPs as the top row instead, it should be nicer to use.
>>
And we're bumped out again.
>>
I feel retarded trying to figure out how transistors work.

What happens to output when bias voltage is increased? Specifically in CE

What purpose does the emitter resistor serve in a CE amp. Stabilizing cc?

Also any good videos on transistors would be appreciated I have been scouring the net for like two weeks and I still don't get it.
>>
>>1277372
For first one I am referring to output voltage rather than current.
>>
>>1277372
>What happens to output when bias voltage is increased? Specifically in CE
Assuming all this talk is about common emitter amplifiers. You should have DC-blocking capacitors anyway, but without them it will probably just offset your output voltage and perhaps introduce some nonlinearity and change the gain a little.

>What purpose does the emitter resistor serve in a CE amp. Stabilizing cc?
I think the emitter resistor serves to keep the output as the value of a voltage divider with a voltage approximately in between the voltage rails, otherwise the output signal would keep getting grounded.

I'd personally just mess about in spice or IRL with a few transistors and a scope, testing what resistors change what and how you can get different results. It's more than any book alone can teach you, but a combination of both is recommended.
>>
>>1277349
Thanks, I figure out how to use the 4 inputs so I get a nice output!
>>
I want a motor to power up around 10 seconds AFTER I plug in the power. Should I use a 555 timer or some kind of capacitor timer?
>>
File: DSC_0032.jpg (610 KB, 2320x1568)
610 KB
610 KB JPG
Is it a bad idea to set up my MOSFETs like this? I planned on building >>1277349
Thanks in advance!
>>
>>1277372
Say it with me:
The emitter resistor is used for thermal stability.

For reference:
https://electronics.stackexchange.com/questions/60696/emitter-resistance-in-bjt-amplifier
>>
>>1277409
Couldn't you buy a DIP with the same configuration?
>>
>>1277413
I honestly don't know. What am I supposed to search for? I didn't find anything so far.

But for now I'll work with what I got.
>>
>>1277409
If they are running hot the heat from all of them could accumulate
>>
>>1277372
> What purpose does the emitter resistor serve in a CE amp.
It provides negative feedback, stabilising the overall gain.

The emitter voltage will be the base voltage minus ~0.65V. The emitter resistor makes the emitter voltage proportional to the emitter current. Between these two, you end up with the emitter current proportional to the base voltage minus ~0.65V (and independent of the transistor's current gain, which varies with temperature and process variation).

A common-emitter amplifier without any feedback is prone to thermal runaway. If the temperature increases, the transistor's current gain increases, which increases the collector current, which increases the power dissipation, which increases the temperature, ...
>>
>>1277418
L293D
>>
File: 5019323_sa.jpg (39 KB, 500x356)
39 KB
39 KB JPG
So I'm using a cheap BT audio module to recieve audio from my phone and feed it in to my cassette deck.
All this is built in to the insides of a cheap tape adapter that has a wire running from inside it for power.

How would /diy/ provide 3.6v and less than 55mA to a chip inside a cassette? Rechargable battery? Harness the cassette's spinning power with a motor?
>>
>>1277777
1 cell lipo battery
https://m.aliexpress.com/s/item/32798926122.html
>>
>>1277409
The tabs are likely live. Be sure they're electrically insulated from one another, or at least can't touch.

>>1277401
>a motor
You need to specify voltage and current when asking questions like this.
>>
NEW THREAD

>>1277831
>>1277831
>>1277831

NEW THREAD
>>
>>1277777
>Harness the cassette's spinning power with a motor?
This isn't a bad idea, if you can get the mechanicals to fit. Try something like a phone vibrator motor with the weight removed as your generator. Have a look at the solutions here:
https://en.wikipedia.org/wiki/Energy_harvesting
But they're probably all too low power for 55mA, perhaps with exception for putting a few magnets on the rotors and surrounding them with coils of very fine enamelled "magnet wire", making sure to position them to get a change in flux as the rotor moves.
If it does work, try to sell them or at least provide plans, it's a cool project.
>>
>>1277777
The hub rotors are designed to slip to prevent the tape from breaking. Not much power to be had here.
The embedded LiPo sounds about right, though you won't necessarily get much runtime out of anything you could fit into a cassette shell.
also checked
>>
>>1277827
5v 200mA. I just used an arduino (sorry)
>>
>>1277834
I was thinking something more along the line of a CD drive spindle or a laptop hard disk, but they spin at very high PRMs and I'm not sure the speed of the cassette deck would generate a current.
>>1277844
Oh geez, I knew they slipped but I didn't think a small dynamo would generate enough resistance.
I could add a connector to charge the cassettes and have 2 or 3 and have it charge over a cigarrette lighter connector or something. But that sorta beats the purpouse of the project.
>>
File: voltage.png (46 KB, 603x384)
46 KB
46 KB PNG
Ive got a question. Will pic related detect zero in 230AC home voltage? if yes how will the voltage betewen A and B look? Is it going to be spike to certain value when zero crossing and then back to zero or is it going to be @ certain value untill zero crossing and sharp drop to 0V followed by come back to certain value?
>>
>>1274633
use external source of power




Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.