[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] [Search] [Home]
Board
Settings Home
/diy/ - Do It Yourself



Thread archived.
You cannot reply anymore.



old thread: >>1413287

>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.

>Books?
Beginner:
Forrest Mims III, Getting Started in Electronics
Charles Platt, Make: Electronics
Michael Jay Geier, How to Diagnose & Fix Everything Electronic

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

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

>YouTube?
mjlorton
BigClive
paceworldwide
eevblog
EcProjects
greatscottlab
AfroTechMods
Photonvids
sdgelectronics
TheSignalPathBlog

>Project/idea websites?
http://adafruit.com
http://instructables.com/tag/type-id/category-technology/
http://makezine.com/category/electronics/

>Components/equipment sources?
Mouser, Digi-Key, Arrow, Newark are global full-line distributors with small/no minimum order.
RS Components (Europe)
eBay/AliExpress sellers, especially good for component assortments/sample kits (caveat emptor)
Your local independent retail electronics distributors
ladyada.net/library/procure/hobbyist.html

>Circuit simulators?
This mostly comes down to personal preference. These are the most common ones though:
LTSpice
CircuitJS (quick, dirty, interactive)
NI Multisim
CircuitLab
iCircuit for Macs

>PCB layout software?
KiCAD (recommended), why use anything else

>My circuit doesn't work. Halp?
Check wiring, soldering, part pinouts, and board artwork if applicable, then post schematic. Supply ALL relevant info and component values when asking for help.
>Li+/LiPo batteries
Read this fine resource first: https://www.robotshop.com/media/files/pdf/hyperion-g5-50c-3s-1100mah-lipo-battery-User-Guide.pdf
>I have junk, what do?
Take it to the recycler.
>>
>>1416580

OP's pic is from pic related, while powering my computer and monitor. Contrary to the advice I was given here in a previous thread, it is a pretty decent sine wave.
>>
>>1416584
Looks fine to me.
>>
Good hobbyist oscilloscopes?
>>
>>1416685

pic related is good if you can afford it. For less than that I'd consider a USB scope that uses your computer, but be extremely careful that you don't damage your laptop or PC; do some reading about the risky methods and follow the rules.
>>
>>1416703

I have one of those but I don't know how to display both the zoom and normal at the same time, as shown in this pic. Anybody know?
>>
>>1416708

press the button circled in red.
>>
File: WHS40D.png (356 KB, 1560x1098)
356 KB
356 KB PNG
I have this soldering station that uses 3,5mm tips, of which just about none exist, that are secured by a shitty set screw.
It's 18V, 40W, 3 pin connector (DIN so not a problem to work around). Any chance there's a suitable chink replacement handle for it with a good tip selection?
>>
Why does amp efficiency get better when the time spent conducting goes down? i.e class A vs B, or is this just some perverse thought that you have 100% efficiency when your amplifier is off? Does a class A actually waste more power than AB?
>>
>>1416685
sds1104x-e
DS1054Z is deprecated
>>
File: amp22.gif (8 KB, 400x308)
8 KB
8 KB GIF
>>1416797
Pic related is a simplistic class A amplifier. Let's assume that the transformer's ratio is 1:1 and that you want to feed +1A/-1A current to the speaker. Since your transistor has to remain conductive even at the negative peaks, you have to set the no-signal collector current to at least 1A, preferably somewhat higher. The collector current will then be close to 0A at the negative peaks and a bit over 2A at the positive peaks. This 1A current and the associated power dissipation does nothing productive, it's just needed to make your circuit work. This obviously hurts your efficiency.
A class B amplifier, on the other hand, draws only as much current as your load needs. A simple class B equivalent for pic related would use dual supplies with the same total voltage as the class A circuit, but drawing only +1A/-1A during the peaks. And the quiescent current / quiescent power would be ideally zero.

This was a rather hand-wavy explanation. If you want to play with math, you can replace the transformer/speaker with a resistor and calculate the total input energy vs. the output's AC component's total energy over one sine cycle.
>>
>>1416818
>sds1104x-e

that's a better scope for a higher price.

do you even know what deprecated means?
>>
>>1416685
oscilloscope is a crutch for the weak and incompetent. a true electronics engineers needs only his mind and a multimeter
>>
What’s a good way to make me not retarded when it comes to breadboards? I have no problem making circuits out of aligator clipped jumper wires and metal coins though
>>
>>1416872
multimeter is a crutch for the weak and incompetent. a true electronics engineers needs only his mind.
>>
>>1416874
>I have no problem making circuits out of aligator clipped jumper wires and metal coins though

pics pls.

and real guys call them gator clips
>>
So what do I put my project on? I really hate just dragging it around and having 10 components just hanging on wires.
>>
>>1416891

I just buy everything I need
>>
>>1416903
>So what do I put my project on? I really hate just dragging it around and having 10 components just hanging on wires.

1. learn how to reply: you click on the number of the post, like I clicked on your 1416903

2. post a photo of this mess that you are working with
>>
File: circuit.png (7 KB, 323x244)
7 KB
7 KB PNG
How do I analyse this? I'm stumped.
>>
>>1416967
looks to me like all the gates are floating at start until the power source is connected at which time the gate net will charge to some random voltage determined by the ratio of the p and n capacitances. that or later noise will eventually pull the voltage high enough to activate the p or n channel fet. if the threshold voltages for the left fets are low enough that both conduct at a given voltage then i think they would enter a stable mutual on state and waste some power. if the threshold voltages for both are high enough that they can't both be on at the same time then it'll stabilize at some random high or low voltage where the left fets don't conduct. the right fets do whatever the left pair does. i have no idea what the purpose of this is but either it's stupid or i am.
>>
I have a bunch of capacitators, resistors and dioes, is there some way to build a voltage and current regulator from it?

I have a dynamo on my bike and i would like to use it to charge my phone, but it supplies different voltage and current based on how fast it spins.
So how would i go about making a circuit that limit the current and voltage to say 5V and 2A and tries to prevent it falling under those values if possible?
>>
>>1417145
it`s a test question from my electronics class. I managed to solve most shit but this really got my almonds. It gives the Vt values and says I`m suposed to consider them all in saturation mode for the analysis.
>>
File: image001.png (8 KB, 328x144)
8 KB
8 KB PNG
>>1417252
You can't fix voltage and current at the same time. "5V 2A" means "5V and [up to] 2A". Preventing it from falling under is hard (boost regulator) but making sure it doesn't go over is easier.

It's possible to make (shit) regulated supplies with a certain type of diode (Zener), but most regulated supplies use a feedback system and an active control element like a transistor. You can build it yourself, but it'll be easier to just pick up a 7805 linear regulator desu.
>>
>>1417275
What is the given value of Vt?
>>
>>1417276
What is that triangle thing? A logical gate? Can you explain more about it? I do have a bunch of transistors and mosfets too.
>>
>>1417304
>What is that triangle thing? A logical gate? Can you explain more about it? I do have a bunch of transistors and mosfets too.

where do people like you come from? are you a poorly programmed alien? You have a pile of transistors and you've never heard of an op-amp.
>>
>>1417294
Vt(n)=1 Vt(p)= -1. I think that as the Vgs in Q1=0 you can find the current in it using the Id formula, and it's the same current through Q3, with that you get VGS3 and as VS3=0 you get VG3 which is equal to VD3=VG(all) and then it goes from there. I think that is correct? he didn't give a correct answer (it's an old test, I'm preparing for another one.)
>>
>>1417304
Op amp
>>
File: 0e9.jpg (27 KB, 600x600)
27 KB
27 KB JPG
>>1417327
comparator
>>
>>1417329
>comparator

LINEAR REGULATOR PEPE
>>
>>1417276
The problem is that if the current or voltage is too low the phone will stop charging and quickly be destroyed since the current won't go into the battery
>>
>>1417341
Nevermind i looked at a store and voltage regulators are dirt cheap, like 50 cents a piece. I will just get a bunch of those
>>
>>1417366
wait why does it needs capacitators? If the output voltage is lower then it never dips below the regulator voltage. Fuck that, i'll skip adding them, probably just the store trying to jew me over.
>>
>>1417381
output capacitors are for filtering and improving regulator stability, not to jew you out of 5 cents
>>
>>/g/66580008
This
>>
>>1417401

>>>/g/66580008

no, just no
>>
>>1417388
>regulator stability
If the regulator is for 5V and i don't use the caps, what ranges are we talking around here? .1V?
>>
>>1417421
The worst case is full 0 to input voltage oscillation.
It depends on loading and what exact regulator you're using. Some regulators work reasonably well without capacitors, some don't.
>>
>>1417461
So out of the 3 used normally, which one is the most important?
>>
If this isn't the right place to ask can you direct me?
I'm hoping one of you guys might know a thing or two about this
I suspect that the backlight of my led TV has gone out. I was playing a game when the screen went black with the sound playing. If I moved closer I could faintly see options through the black so I switch it from game to vivid mode and the picture came back as if nothing was wrong. I carried on using for the day. So comes today and the screen is black with sound and changing the picture mode doesn't work.
What's different now is I can notice then screen flicker sometimes(note this flicker was with the black screen as before were I could faintly see the options for me to change picture mode)
So is this a backlight failure? Can I fix it myself?
>>
Ok... now that I got a halfway decent regular multimeter, I’m onto the clamp meter idea. But I really want one that reads DC Amps as well. Thinking about automotive in this case as well as so much of the stuff that runs off an AC adapter.

There’s one on Amazon for around $45 but like 80% of the reviews are “This thing is great for the money!” and the other 20% are “This doesn’t work at all!”

I’m really thinking about HF on this one but goddamn I hate spending over $30 on anything electronic there because of the 90 day warranty. This 600A AC/DC model (2nd from right in pic) has great reviews though. Home Depot’s website has a 600A Klein or Extech for around $120, and I saw a 400A Extech for $85.

Any thoughts?
>>
>>1417602
This one honestly doesn’t look bad for the price.

Lowe’s carries a brand called Southwire and it seems line their answer to Klein with overpriced screwdrivers. They have a 400A AC/DC clamp meter for $79 that I was looking at in the store today but the reviews weren’t too strong.
>>
>>1416580
Can I recommend Brian Douglas's youtube channel? He covers a lot of control theory which is important for robotics/drones/etc. Also he's doing a series on the MATLAB youtube channel on PID control which is nice to know about. Dunno if it's too "theoretical" for diy projects howefere
>>
File: 1456883425340267[1].jpg (64 KB, 700x376)
64 KB
64 KB JPG
>>1417602
If you're not doing >100A consider Uni-T UT210E. Small, reasonably cheap, does DC current down to 1mA resolution.
>>
File: 1397310818119.png (56 KB, 841x884)
56 KB
56 KB PNG
>>1416872
A true electronics genius only needs the power of his mind to measure waveforms, solder connections and supply power to his circuits. Physical tools are for the weak.
>>
So I think I accidentally made an EMP, is there some equipment I can build to test this? So I charged up a capacitor to about 10 KV and it started arcing repeatedly and an (unconnected) power supply was found to no longer be working. The power supply was at minimum 2 feet away and has long cables.
>>
what was the name of that microcontroler that had an LCD screen on top? it was going for about 10 bux on ali.
>>
>>1417793
ESP32
>>
File: 20180703_054018.jpg (2.43 MB, 1917x1920)
2.43 MB
2.43 MB JPG
What are these components named?

The bottom right looks like a green LED. But I'm sure it's not. The large two rectangles in the top right look just like MLCCs but I've never seen them with text printed on the face. The top left I have no idea.
>>
>>1417819
bottom right is a resettable PTC fuse. top rights appear to be high voltage capacitors by the reference designators. left is a coupled coil of some sort, like a transformer.
>>
>>1417663
How hard would be to make my own if i only want it to measure dc amps?
>>
Putting together a custom shield for a really simple arduino project, can I connect both grounds together? What little understanding of this stuff says it shouldn't matter and I have no idea how to make sure (aside from checking but I haven't ordered any components yet). Project is a 9 button arcade controller, so my shield is merely routing all the pins to a bunch of JST-XH connectors for the buttons/LEDs.
>>
>>1417865
Ideally you want to avoid ground loops. That is, ground wires are joined together at 2 or more places. Ground loops are also knows as antennae. That said, at arduino speeds and voltages it probably won't cause a problem.
>>
>>1417873
If it's a best practices thing at this level I can do it right. I wanted to join them to make the routing a bit prettier.
>>
>>1417875
what exactly do you mean by "both grounds"? all circuits have to have a common ground, generally from your power supply. the shield is getting its ground from the arduino
>>
I want to connect raspberry pi directly to my car battery contacts. Now the car battery is only about 13V but for example during starting it can jump up to 1000A, but that should not be a problem right since the voltage is consisent?
The rpi needs 5V 2.5A to operate so if i connect the ground to the battery - and plus to the battery plus and put also 3.2 ohm (8/2.5 = 3.2) resistor in series on the plus wire it should work right?
But 3.2ohms seems small as fuck.. am i missing something?
>>
>>1417880
get a cheap down converter that always has 5V, not only at 2.5A
>>
>>1417888
Would it would without it though if we assume the voltage stays at 13? I just cant believe the resistor i need is so small.
>>
>>1417891
What you are proposing won't work. Don't ask me to explain. Just buy a cheap converter that has a USB you can plug your RPi into.
>>
File: this.png (10 KB, 1105x555)
10 KB
10 KB PNG
>>1417865
if you are doing this, don`t. you have to use a single ground.
>>
>>1417899
>Don't ask me to explain
If you don't know just say so.
>>
File: raspi-amps.png (48 KB, 610x400)
48 KB
48 KB PNG
>>1417891
calculate the voltage you would feed your raspi with after 13V and 3.2Ω
source: www.raspberrypi.org/help/faqs/
>>
>>1417880
The raspberry pi will pull anywhere from 100mA to 2.5A, you can't just assume a complex digital circuit will pull the same current, that would be stupid. We don't use resistors for this purpose for this exact reason. We can use a linear regulator, which maintains a constant voltage on the output regardless of output current, but this is wasteful and gets as hot as your resistor would, or a maximum of 20W in this case. 20W would blow any common resistor or linear regulator to pieces quite quickly. So instead we use a switching regulator, which uses an inductor to step down voltage without dissipating power in a resistive component. Something like pic related would work fine:
https://www.aliexpress.com/item/1488/32738093688.html
And it's less than a dollar, though its 3A rating might be a little suspect. It's not isolated though, so don't do anything with audio without an isolation transformer (car audio grounds are at ~6V relative to the 0V terminal of the battery). Might want to put a diode before the input capacitor too, but it probably won't matter. You may also be able to find a similar model but without the USB output, just some solder pads.

>>1417901
Don't be a shitlord.
>>
>>1417901
>If you don't know just say so.

so are you trolling or just bone headed?
>>
File: this.png (9 KB, 1105x555)
9 KB
9 KB PNG
>>1417900
do this
>>1417880
why in the contacts? aren`t there connections inside the dashboard you could use? and if the power oscilates that much in the battery it`s stupid to plug a MCU or computer directly into it.
>>
>>1417901
If you think you are right then just do it. Post results so I can laugh some more.
>>
>>1417904
>>1417901
Oh and linear/switching regulators have plenty of ripple rejection, meaning a fluctuation of the power rails will result in a very attenuated fluctuation at the output. So you probably won't have to worry about the thing dropping to 9V or so when you crank it. Also they don't pull 1000A, probably closer to 150A.
>>
>>1417904
>linear regulator
I actually have one of those.
20W shouldn't be too bad. I mean my computer does like 120W under load and that it can be handled easily by the shitty stock fan. Si if i slap a cpu sized aluminium cooler on the converter i should be golden.

I will order the one from ali, but that takes almost two months to arrive here, so i will use the linear one instead for now
>>
File: 1503137966073.gif (655 KB, 460x426)
655 KB
655 KB GIF
>>1417915
your computer is also using a switching power supply, not a linear regulator. in fact, probably several switching power supplies
don't forget to report back tho
>>
Anyone here ever got their project featured on hackaday? Or are we all dumb fucks?
>>
>>1417917
Doesn't change the fact that the CPU is generating lots watts of heat which the cooler on it has to eat
I will test it out with a raspberry zero first so if it melts i will be down only couple of bucks
>>
>>1417918
At least two Hackaday articles have covered shit which was first posted here.
Not that it makes you or me any smarter.
>>
>>1417332
>confusing the component with the circuit

Anon I don't even
>>
>>1417930
context is everything

>>1417919
good point.
one more tip, be advised that TO-220 packages' tabs are usually connected to the middle lead. fortunately for you, the 78xx middle lead is GND so you could just screw the device right to the body with a lock washer and a reasonably light touch
>>
>>1417663
I saw a Uni-T on Amazon that looked decent but was low amps. I’m not exactly sure what I will run into with cars and household stuff. I’m gussing I don’t need the 1000A meter for that but I don’t want to go so small that I need to upgrade in the middle of a project.
>>
>>1417602
>DC Amps
Shunt+voltmeter?
>>
>>1417957
That sounds too complicated for me. I think I want a clamp meter so when the YouTube video says “Now take out your clamp meter and see what the reading is,” I can take out my clamp meter and try not to make things worse.
>>
File: Capture.jpg (14 KB, 335x72)
14 KB
14 KB JPG
ESP32 internet radio retard anon here. I thought I had ordered all the parts I need, but now I'm not sure anymore.
The 16x2 display is a TMBC16265G with a S6A0069 controller. I can't find a proper datasheet for the LCD - only the pinout. The datasheet for the controller says pic related, but I'm not exactly sure what V5 is. The VDD pin is currently fed with 5V.
Is the 3.3V on the IO pins enough, or do I need a level shifter?
>>
>>1417959
Well, up to you. The way it works is really simple, literally just ohm's law. On the other hand I guess it does affect your circuit more than a clamp meter would (?), and you need to include it in the circuit before taking the measurement.
But, 5€ instead of 100€... https://www.amazon.de/75mV-aktuell-Shunt-Widerstand-meter/dp/B008QV0GKO/ref=pd_sbs_328_3?_encoding=UTF8&pd_rd_i=B008QV0GKO&pd_rd_r=cc8fad66-7951-11e8-ab87-4fddd0aceb00&pd_rd_w=2LZDD&pd_rd_wg=xjONt&pf_rd_i=desktop-dp-sims&pf_rd_m=A3JWKAKR8XB7XF&pf_rd_p=1662399603728955198&pf_rd_r=09AYH5XKS1GDR2K8CRNM&pf_rd_s=desktop-dp-sims&pf_rd_t=40701&psc=1&refRID=09AYH5XKS1GDR2K8CRNM
>>
Are the C1 capacitors in this circuit polarized or bipolar? I'm a bit confused because the C2 capacitor explicitly has the "+" next to it but they don't and I don't want to get the wrong kind.
>>
>>1418092

I'm not sure I can answer anything you asked, but this link:

http://www.cloverdisplay.com/pdf/S6A0069.pdf

has pic related on page 24. V5 might be GND and it might be something else.

as for level shifting, page 31 is for Vdd = 5 volts, and Input Voltage (2) would need level shifting; what the two Input Voltages mean is probably covered in the spec somewhere.
>>
>>1418101
At 47uF they're going to be polarized and R1 side would be positive.
The two ICs on the left side have their pin 1 left floating and it doesn't look intentional.
>>
>>1418118
Oh shit, I guess this doesn't even matter along with a boatload of stuff in that data sheet. I guess I just need to know what the relationship between VDD and DB0-7 is. Is the data signal supposed to be the same as VDD? So if I'm running VDD at 3V data bits are 3V, and if VDD is 5V, data bits are 5V?
Either way I might as well order a level shifter though as they are pennies.
>>
>>1417918
lel do you consider hackaday prestigious or something?
>>
>>1418172
>http://www.cloverdisplay.com/pdf/S6A0069.pdf

After looking at page 31 again, I think it is clearly saying that with Vdd 5v, Input Voltage for all but OSC1 can be 3 volt compatible, meaning DB0-7 can handle 3v or 5 volt inputs.

likewise for Output Voltage (1) which is for DB0-7, that is also 3 volt compatible for Vdd 5 v.
>>
>>1418186
i mean it's a pretty big website. Imagine being able to write "featured on hackaday" or "featured on magpi" on your cv
>>
>>1418200
I just write "featured on 4chan".
>>
>>1418191
Ah, think I'm starting to get it now. Why isn't the maximum output voltage (for VOH) listed? Couldn't this be higher than 3.3V?
>>
>>1418236

The input maximum voltages are specified because they come from elsewhere, and some other part might send in a voltage that is too high and damage this chip.

The output maximum voltage is implicitly Vdd, which has a max value specified somewhere else, but this chip cannot exceed its maximum output voltage if you supply the proper Vdd. In other words, there's no need to specify it in two places. Output high voltages are only concerned with being high enough, in this case 2.4 volts, whether Vdd is 3v or 5 volts or whatever.

Now if you are concerned about the ESP being hit with a 5 volt signal then that is a valid concern, and if this chip has Vdd 5 volts then you need to use a level shifter. If the line is a one-way line, meaning it is always an output, then you can use a voltage divider. If it is bi-directional then you need a proper level shifter.
>>
>>1418202
doesn't have the same ring to it. Maybe if you write "featured on message board with millions of users"
>>
>>1418242
Thanks for clearing it up. The R/W pin is connected to ground so it's always in write mode, so it shouldn't actually output anything ever.
>>
Fucking shit I need some desoldering tips. I'm trying to change the switches on a keyboard, and it came with leds, and those are killing me.
I'm using a solder sucker and I have a like 90% success rate with switch solder joints, but each and every led joint is a fucking battle. I think it's because of the smaller thermal mass of the tiny led joints, in the instant it takes me to plomp down the sucker and release it some solder already solidifies and leaves a bridge. As a result I don't know what the fuck to do.

I do have copper wick/braid, but I think it has autism or something. Sometimes it wicks up a bit of solder, but it's never enough. I've tried both going raw and fucking drowning it with my flux pen, didn't make much difference. I have not once been able to "finish" with the wick after starting with the pump: every success so far has involved simply adding back a ton more solder then going for another attempt with the sucker. Then once I've got one leg painstakingly free, I hold the iron to the other leg and pull, because even just getting one leg properly free is a nightmare.

What am I doing wrong? I don't really want to cave and buy a desoldering iron because I don't plan on doing much more desoldering in the future, and also I am clearly doing something not right and things could be much easier. That said if there's a good one for maybe £30 or less I could consider it.
>>
>>1418202
I'm sure that gets a bitch-ton of callbacks

>>1418236
maximum input voltage is generally dependent on Vdd, because bad things tend to happen when inputs go outside the supply rails. in addition there may be an absolute limit to input voltages which is dependent on manufacturing process. both apply

>>1418316
>not plomping down the sucker before heating
shiggy
is your desoldering tip burnt/chewed up? if so, trim it down
are you heating the joint through the wick? if not, do that
are you using an appropriate iron temperature for what is almost certainly lead-free solder?
last time I replaced a keyswitch, I heated both LED leads at once and pulled the LED out by itself with some large long-nose pliers. the keyswitch pads were large enough in diameter that I could suck one side mostly free. sometimes, on through-hole components, it helps to rock them out, heating one side then the other, pulling gently on them the whole time
>>
>>1418329
>not plomping down the sucker before heating
Wait what? I can't heat the joint with my iron if I have the sucker on it.
>are you using an appropriate iron temperature for what is almost certainly lead-free solder?
I started with 350°C, then I lowered it to 320 because I was afraid of just fucking my PCB due to the time I'd spend trying to heat it. I didn't notice much difference if at all - it's a chink keyboard (i.e. bough and shipped directly from china) so maybe it's leaded, I guess.
> I heated both LED leads at once and pulled the LED out by itself with some large long-nose pliers
That could certainly be something, actually. I can probably start by trying to suck off as much as possible, then just heating both at once. Certainly worth a try.
>>
File: diffpair.png (7 KB, 282x238)
7 KB
7 KB PNG
what's the best way to get differentially wired transistors "matched" without going through the trouble of hand-picking through a huge batch or trimming? Precision resistors on the emitters? Local negative feedback? Does anything work?
>>
File: 1530379765324.png (195 KB, 850x1100)
195 KB
195 KB PNG
>>1418365
the sucker tips are somewhat heat-resistant. it should be possible for you to put the sucker tip and the iron tip within a few millimeters on opposite sides of the joint so the sucker tip is on one edge pointing at the joint, then pull the soldering tip away and lean the sucker over so the tip is flush while pushing the sucker button
>cheap
I bet it's lead free, either way, you usually need more heat for desoldering operations. don't worry about sucking until the leads are out. then you can iron on one side, suck on the other to clear the hole

>>1418369
buy a pair in one package
>>
>>1418377
>buy a pair in one package
options for through-hole parts are slim and expensive as fuck. I also need more than just a pair to be matched.
>>
>>1418385
why not SMD then?
>>
>>1418385
TH components are out of fashion grandpa
>>
File: schematic.png (28 KB, 1275x720)
28 KB
28 KB PNG
Is there any problem with using one coil of a toroidal transformer or a choke as an inductor for a switching power supply?
Pic related is my schematic btw, basing component values off of one of those online calculators, the output is 450V for powering dekatron tubes.
>>
>>1418396
It will be a source of noise. You need a shielded inductor for a SMPS.
>>
>>1418397
Is there a way I can shield them? I would just buy some, but it's much cheaper to use what I rip out of old power supplies instead of buying new inductors for $0.50-$1 each, on top of having to wait for them.
>>
>>1418399
If you're ripping shit out of old power supplies, why not rip the wire off the transformer and re-wire it with only one winding?
>>
>>1418406
Isn't that still going to be unshielded?
>>
>>1418396
>>1418412
he's being silly. you just have to connect the terminals of the unused winding together to keep the noise inside.
>>
>>1418385
they don't specify device-to-device matching, but you could take your chances with consecutive units from a cut tape, which halves the units needed to match, fwtw
if that's not good enough, then you might need to build an automated jig for measuring the parameters of interest to you. whatcha makin' anyway and what parameters do you need to be matched?
>through-hole
pops, I am disappoint

>>1418396
pin 1 *probably* isn't rated for 400V
you're going to need the secondary of that transformer (assuming it has 10:1 or larger turns ratio) or a 600V MOSFET to make that kind of voltage. also probably a higher-impedance feedback network
>>
File: schematic.png (30 KB, 1538x536)
30 KB
30 KB PNG
>>1418451
You're right, it's not. I put a 500V mosfet in based on the example in the datasheet, but I have no confidence that I did it correctly.
>>
>https://oshpark.com/shared_projects/qZQfIT9V
I sincerely hope nobody ITT is responsible for this travesty
>>
>>1418474
ew
>>
>>1417517

replacing the backlight on a TV is super duper easy after you've done it for the 20-th time. the first 10 times, you'll damage the screen beyond repair, and the next 10 times will result in various levels of breakage. but on your 20-th attempt, SUCCESS!

problem is, the backlight itself is the problem only 50% of the time, the other likely culprit is the high-voltage circuit (about 1000V) that powers the lamps. to determine which is which, i use a 1000V circuit i pulled out of an old scanner to test the lamps.
>>
>>1418466
SwC goes on the other side of the inductor, otherwise you risk blowing the IC and/or the FET's gate

>>1418474
jesus fucking wew

>>1418510
absolute namdam
>>
>>1418466
Generally speaking, getting 450V from 12V with a basic step-up switcher is a bad idea. It's rather likely that your regulator simply can't produce that high output.
Also, like the other anon said, SwC should go to the inductor's other end. There's also a junction missing near Ipk and 1N5819 can't handle 450V. Most likely the mosfet pull-down resistor has to be smaller, too. Replacing it with a buffer would probably be a good idea.
>>
>>1418573
I know it's generally not a good idea, but for this application I figured it might be okay since it pretty much doesn't matter how noisy the output is. That being said, I'm pretty much pulling that out of my ass because I have no idea what I'm doing with power supply design, I just got the cheapest chip and simplest circuit I could find.
If this sort of design won't work, what should I look to use instead?
>>
File: 1-1366309322.png (8 KB, 373x178)
8 KB
8 KB PNG
>>1418577
>If this sort of design won't work, what should I look to use instead?
One which uses a transformer.
In the simplest case this would be a normal mains transformer, like in pic related. If needed, use a diode doubler or tripler to get the voltage you need. I doubt your tube needs particularly well-regulated supply.
>>
>>1418577
2nd hit on ddg http://www.vecoven.com/elec/hv_supply/hv.html

>>1418584
>tip
mild lel
>>
>>1418586
that design seems neat enough
I'm assuming I can use a BC337, BC327, 1N4007, and any diode that can handle the voltage instead of the specific parts they specify without any problems, right?
>>
>>1418597
not just any diode, you want especially fast ones. use an ultrafast recovery type e.g. UF4007 for best results, its reverse recovery time is 100x faster than a 1N4007, which means less loss, therefore less heat and more voltage. whatever's on hand should be fine for the driver transistors. use a measured/specified inductor if you want any certainty of success
>>
>>1418600
the fastest i have is FR107s, still about 10x slower than those UF4007s
>>
>>1418600
>>1418602
oh and I also have a pair of surface mount 220uH inductors so that's no problem
>>
>>1418603
>no problem
calculate peak current and see if your L can handle that without saturating
>>
File: 1445823559157.jpg (20 KB, 500x313)
20 KB
20 KB JPG
>>1418600
>ultra fast diodes
>555 timer
>>
File: 2.jpg (98 KB, 1558x654)
98 KB
98 KB JPG
This question is from the Art of Electronics, and I'm having a difficult time with it.
I need to add a collector resistor to this circuit, because the output voltage (the voltage across the emitter resistance) is 5 V, and then if I were to operate this in saturation, the V_CE would be something like 0.3 V. Therefore I need a collector resistor to drop 15-5.3 - 9.7 V.

I don't know the collector current, however, with my assumption of using the max emitter current, I can find it if I had a specific beta. The problem with this, I think, is that poor designs use beta as a determining parameter according to AoE. So, how should I choose a collector resistance?

Additionally, what is meant by
>Choose your resistor values so that the output voltage doesn’t drop more than 5% under full load.
Full load isn't defined in the question is it?
>>
>>1418749
If you add a collector resistor then your 5v comes from the voltage divider you created and nothing to do with the transistor it might as well not be there.
Load means the voltage at the emitter should be 5v @ 0A, and no less than 4.75(?)V @ 25mA.
The entire purpose of this exercise is to avoid saturation region isn't it? The transistor needs to drop the 10v so it can't be on fully
>>
>>1418755
>If you add a collector resistor then your 5v comes from the voltage divider you created and nothing to do with the transistor it might as well not be there.
True.
>Load means the voltage at the emitter should be 5v @ 0A
I don't know how one would do that.
>The entire purpose of this exercise is to avoid saturation region isn't it?
I honestly don't know.
>The transistor needs to drop the 10v so it can't be on fully
That makes more sense. I need to go back a few pages, I think, because I don't remember how to operate the transistor in the linear region.
Thanks.
>>
File: 2.png (28 KB, 1530x644)
28 KB
28 KB PNG
>>1418759
idk, maybe this would work? I just made the voltage divider give 5.6V at V_b, so that technically V_e should be 5V. Higher emitter resistance puts me closer to 5V on the emitter, however, I'm not sure if that's bad or good.
>>
>>1418774
looking good i would say, exactly you have to bump the Vb to compensate your 0.7v drop Vb to Ve right? or 0.6 you said whatever fine.
so now you are operating as a voltage follower, the problem is that as your R3 decreases ... uh... something happens i think it draws more current through the base or something? fuck i haven't done this in a long time, look, you are doing this in spice right so set up a parameter sweep for R3, change its resistance from 0 to 1M or whatever and plot the result, your voltage will be through the floor to start then as the resistance increases it will tend towards Vb (-0.7) and level off as the resistance exceeds a value that would influence the voltage divider. because you are kind of adding an extra path into the bottom of your voltage divider right? so your voltage divider isn't r1 vs r2, its r1 vs r2 in parallel with r3 kind of. right?
>>
>>1418774

1M on the emitter is insane. if you want a 5% drop at most when you connect the load, all you have to do is make sure the emitter resistor is 20 times smaller than the load load. so, if the load is gonna suck 25mA at most, then you make the emitter resistor suck 500mA. that's a 10 ohmo resistor, which is a long long way from 1,000,000 ohmos.
>>
File: 1523687445893.gif (1 MB, 498x213)
1 MB
1 MB GIF
>>1418602
FRs might do if you don't run the 34063 at the fast end of its range, which with all that inductance you probably wouldn't

>>1418693
>can't into replying
>replies
>>
>>1418810
i'm just gonna grab some UF4007 equivalents from my LES for 60 cents each, no reason not to
>>
File: IMG_20180704_163157.jpg (2.07 MB, 4136x2198)
2.07 MB
2.07 MB JPG
Sup /ohm/. I gutted a voltage stabilizer (placebo) that was freaking out and cutting off at random and just wired it to be an outlet (need plugs for some low-power shit like USB charger and home networking equipment, nothing too power demanding) by removing the transformer and IC board. This is what I ended up with. The PCB just connects to the sockets; neutral and earth solder straight onto the other side of the PCB. A wire for phase/live come out from under the right side of the PCB and is connected (with heatshrink for safety) to a 5A breaker with a red backlight for the switch, which wires into the wall. The dangling blue and brown wires are where the IC board used to connect to, I just wrapped them up in electrical tape and left it hanging for now

First off, dad's an idiot and wired phase and neutral backwards in my wall so I still get a voltage in the sockets even if the 5A breaker is off, thanks dad. The two M Y S T E R I E S are that the breaker backlight that isn't working anymore, and that when I flip the breaker open I still get 30VAC between what are supposed to be phase and neutral (according to my meter). I plugged in an old junk Ethernet switch and it's working fine, though. Any clue why the light's not working on the breaker or why I get those 30V between live and neutral even when the breaker's open?
>>
>>1418817
>why I get those 30V between live and neutral even when the breaker's open

it's a phantom induced voltage displayed by digital multimeters. if you put a load across the meter, like 1meg, it'll disappear. or measure with an analog meter for vintage precision.
>>
>>1418817
>why the light's not working on the breaker

it's supposed to have 3 connection points, but i only see 2 in the pic: (1) switch contact 1, (2) switch contact 2, and (3) light
>>
>>1418828
Huh. There was never any wire connected to it. Maybe the 3rd connection made contact with a solder point when put in one way and I have it backwards, but that sound really nigger-rigged.
>>
So my chink solder sucker is fucking dying. I don't know what I expected desu, but I do need a replacement.

My question is, what's more worth it: a really high end sucker, or a really low end desoldering iron? I found a sucker that's supposedly super duper amazing puper for £18, I haven't looked into desoldering irons yet but I don't want to spend over £30 or so. Basically I'm probably going to go with the sucker unless somebody tells me "no, if you buy the sucker you it will barely improve on the $1.50 chink one, but if you buy this iron for £35 you will cum rainbows every time you desolder" - I'm not at all looking for expensive stations because I'm only going to need it very occasionally.
>>
>>1418864
Wat. I'm gonna need a diagram here. Or a drawing. or some links.
>>
>>1418864

a desoldering iron reduces the difficulty of the job by 90%. a super duper sucker is still a sucker; all of the difficulties remain intact.
>>
>>1418870
A diagram of what? A solder sucker?

>>1418872
How much should I spend to get a usable desoldering iron? Any recs?
>>
>>1418876
>Any recs?

nope, mine's 20 years old, and no longer for sale. that also means i'm not up to date on prices, coz i havent looked for 20 years.
>>
>>1418880
Fair enough. Anyone else itt?
>>
You know one of these days I'm going to drop my hot iron on my fucking dick
>>
File: Capture.jpg (109 KB, 766x624)
109 KB
109 KB JPG
How would I go about using some sort of switching jack to have a bypass operation in this ring modulator? I would like to have it so that when the carrier is plugged in (labelled audiojack right now) the switch is open, and the ring modulator is being used, but when the jack isn't plugged in, the switch would close, and the audio would bypass this part without any sort of modulation to the signal.

My main issue is that switching jacks really don't make any sense to me; mostly because the documentation is such shit, take this amphenol switching jack's datasheet for example:

https://www.mouser.com/datasheet/2/18/55010544000110_acjx-nv35-x-771264.pdf

Any help / suggestions from you guys would be appreciated.
>>
>>1418934
i dont know what to tell you, the documentation is pretty clear.
take ACJM-NV35-2S
when the jack is out pins 2 and 3 are connected together. when jack is in 3 is connected to jack sleeve and 2 is conneced to nothing.

jack 1 pins 1,2,3 connect to T? pins 2, 1, green wire that used to be connected to T? 1
jack 2 pins 1,2,3 connect to T? pins 1, 2, green wire that used to be connected to T? 2

basically put the pins 2 and 3 in between the input to the transformer and the transformer.

i'm assuming you want to bypass from pin1 on the left to pin 2 on the right, the two wires going in and out of the left and right. i don't know what the fuck the audio jack at the bottom is doing if you want it to do anything.

if you want your hand held then find someone who knows what the fuck a ring modulator is because i dont.
>>
>>1418864
my Edsyn Soldapullt™ has been solid for a couple of decades
honestly fám it might be worth considering learning to into hot air

>>1418934
not happening on the carrier jack without it having an isolated switch, which appear to be pretty rare these days
OR
some active circuitry to switch the output
>>
>>1418797
>fuck i haven't done this in a long time
same, that's why I'm reviewing it. I feel pathetic.
>>1418802
I agree, it's really high. But using your rules of thumb, puts me really far from the design specifications. At this point is a matter of choosing the right transistor?
>>
>>1416580
Can a big brain electrocity individual explain to ooga-booga anon here why inverters fail? Aren't they just diodes and resistors?
Is it just classic wifi router deliberate incompetence of "lol this is a cheap commodity item, why would we bother making it last forever with and extra $3 of aluminum and a $2 fan?"

tl;dr
Why do solar inverters suck so much?
>>
>buying 30 meters of chem-wik comes out to about the same per meter as buying chink wick that may or may not have flux in it
>a single roll of 1,5m chem-wik is £3
>want to make my own flux
>have to buy >500g of rosin for it to be reasonably priced
I'm probably never going need this much, but yet I feel compelled to buy in bulk.
>>
>>1418131
Ok thanks, I'm pretty retarded when it comes to electronics, can anyone hazard a guess as to what they might be connected to? This is the only pic of the internals as far as I'm aware - it does look like it's not connected, but there must also be a lot of vias under the the chip, judging by the diagram in >>1418101
>>
>>1418975
It's rather likely that your mistake here is using Spice. I mean, questions like that tend to have very simple answers, while your results are complicated by Spice's somewhat realistic transistor model.
Typically a problem like that would go so that you assume that the Vbe is constant and then dimension the base voltage divider so that the base current resulting from your maximum output current won't upset the base voltage more than 5%.
If you really want to take the Vbe variation into account, some minimum load certainly helps a lot. 10R is ridiculous, though.
>>
>>1419048
Based on your pic, that 47uF is bullshit and the correct value is 47pF (non-polarized), which makes more sense anyway.
Try to find a more trustworthy schematic.
>>
>>1419037
You can go full negro and dissolve pine resin in alcohol. It's obviously not activated, so the resulting flux won't be that great.
>>
>>1418975
This exercise is supposed to be solved by understanding and calculation, not speculation and blind spice gaming. In my version of AoE it is exercise 2.2 and the concept behind the question is explained in some detail. If this missing in your version?
>>
>>1418973
> honestly fám it might be worth considering learning to into hot air
Does that basically work by blowing hot air at an area of the board and then just pulling the component out while the solder's wet?
>>
>>1418975

when you change the emitter resistor to 10Ω, you have to readjust the voltage divider. the 10Ω is reflected back to the base as 10Ω times the transistor's beta. assuming beta is 100, it looks like a 1000Ω resistor in parallel with R2. that makes the lower resistance 833Ω. so to get 5.7V at the base, the top resistor has to be 1360Ω.

but that's bad design coz it relies too much on the beta of the transistor. to mitigate that, you need to make the 2 resistors on the base much smaller. the smaller the resistors, the less the reflected emitter resistance matters in the voltage divider equation, and the better the % regulation is gonna be.

it seems the whole point of the exercise is to show that voltage regulation using a voltage divider is a fool's errand, which it is.
>>
File: 3164.jpg (31 KB, 640x489)
31 KB
31 KB JPG
Do you know some HP/Tek/others good company portable analog oscilloscopes? Like the Soviet S1-118 (12x21x30 cm, or 4.8" x 8.4" x 12"). I have a HP 1740A scope, but it is too big, really laboratory device.
sorry, i don`t write in english good, it`s not my native language.
>>
>>1419158
Why analog, if you want portability?
>>
>>1419105
Well, I did find a 500g bag of some polski stuff (kalafonia) that is allegedly activated. I assumed it was a meme though. The stuff in that dangerous prototype video seemed to be just regular natural rosin and it worked well?
>>
>>1419161
I've used the natural stuff out of a tree and it works fine, albeit a little smellier.
>>
>>1419160
Analog scope don`t distort signal form (distortions are determined only by the bandwidth, enhance 20 MHz more easy than convert it in digital form without good ADC). Digital scope with similar parameters (transmission of waveform) have a higher cost. And i like old analog devices :)
I can buy this S1-118, but bulid quality and quality some components is terrible.
>>
>>1419162
Compared to the activated stuff or commercial flux?
>>
>>1419104
Ok thanks, I'll post on some forums and see if anyone has anything else but I think unfortunately this is the only diagram out there. It's quite a rare accessory and there aren't any official diagrams from the manufacturer - the best I can probably do is get some higher definition shots of the interior and even that's a stretch because there aren't likely to be a lot of people wiling to open theirs up.
>>
File: equivalent circuit.png (5 KB, 488x314)
5 KB
5 KB PNG
>>1419131
The chapter in the book is about the input and output impedance of an emitter follower circuit. The exercise is about minimizing the drive current under the given 5% drop conditions.

Basic approach in the book (assuming constant B=100 and constant Vbe=0.7V):
Rout = ∆Ve/∆Ie = 250mV/25mA = 10Ω -- this is a dynamic resistance, not a static resistor!
Rin = Rout*B -- this is the dynamic load resistance at the base
R1/R2 = 9.3/5.7 -- this is the divider ratio at no load for Ve = 5V
R1||R2 = 1000Ω -- this is the source resistance of the divider for Rout = 10Ω

Two equations, two unknowns, solve for R1,R2.

Different approach: R3max = inf, R3min = 4.75V/25mA = 190Ω
R1/R2 = 9.3/5.7 -- divider ratio at no load for Ve = 5V
∆Ib = 0,25mA -- no load to max load
∆Vb = 0.25V -- max Vb drop at max load
Ri = ∆Vb/∆Ib = R1||R2 = 1000Ω

Same result as above. The exercise is about understanding, not about replacing a regulator.
>>
File: pdip16.jpg (18 KB, 550x450)
18 KB
18 KB JPG
>>1415716
Suggested to use a HC4028 bcd-to-decimal decoder and after looking at datasheets it seems to really match what i want to do (thanks anon!)
However all the references i found are discontinued products. There are a few things on ebay but the price is outrageous. Where could i buy 5-10 of these in dip format?
Also is there a way to decipher these chip names without looking at each datasheet individually? like TC74HC4028AP, M74HC4028B1R
>>
File: 50518640_p0.png (2.13 MB, 1920x1200)
2.13 MB
2.13 MB PNG
>>1419407
The first 1-2 letters are usually manufacturer, SN for TI is the only one I remember off the top of my head
"74HC" means "High-speed CMOS, using 74xx family pinouts" (as opposed to 74LS etc).
Obviously the 4xxx is the old CMOS designation for your part.
The suffix usually defines the package and temperature/reliability rating, but vary by manufacturer.

https://dmohankumar.wordpress.com/2012/04/24/know-the-meaning-of-transistor-and-ic-codes/
>>
>>1419131
>you need to make the 2 resistors on the base much smaller. the smaller the resistors, the less the reflected emitter resistance matters in the voltage divider equation, and the better the % regulation is gonna be.
Indeed, this fixes it.
>>1419260
>Same result as above. The exercise is about understanding, not about replacing a regulator.
Interesting, I guess I took the exercise too literally. I have no idea what people actually use in circuits, so the authors intent was lost on me.
>>
File: 1527604922356.jpg (4 KB, 125x124)
4 KB
4 KB JPG
>>1416580
you could have waited just a few minutes to post the new thread for a much better chip of the thread

>>1419122
bingo

>>1419407
there are also a few BCD-to-decimal decoders in the 74xx TTL family, usually in the 7440-7448 range, some of which have outputs that can withstand higher voltages (automotive power is dirty). you'll still want to check the datasheets to make sure the devices can handle the voltages you'll be dealing with, or you'll need to regulate it down to whatever they require
the important part is the 74, which almost always tells you you're dealing with a commercial-temperature TTL-ish discrete logic device. the A and B revision markers may have been kind of important in regard to 40xx logic, but I don't remember offhand what they meant, possibly Vdd range or something like that
>>
>>1419441
>you could have waited just a few minutes to post the new thread for a much better chip of the thread

As the OP of this thread, I apologize if I screwed up and admit that I do not regularly read every post to this thread.

Having said that, can you re-phrase what I did wrong? I'd like to be able to say that I always post a "much better chip of the thread", but I can't decipher that comment. Anybody?

And what every happens here in "a few minutes", anyway?
>>
>>1419445
He just wanted a 6581 thread. It's an old sound chip.
>>
>>1419447
>He just wanted a 6581 thread. It's an old sound chip.

Thanks, but now I'm sorry I asked.
>>
>>1419441
> bingo
Does that work for large through-hole components though?
>>
>>1419447
>an old sound chip
>an
I will not be b8d today

>>1419460
like a signal transformer or a soldered-on TO-220 tab? it can, but it can take a minute, and in the case of the TO-220 you'd probably want to work from the component side to get it clear first. it would also work well for 40-pin DIPs and such, if you have a means of reflowing all the leads at once (nozzle, baffles, or just a 2" wide air outlet)
air preheating the board does help for almost any rework operation tho, even if you do combine it with an iron
>>
>>1419491
More like keyboard switches. Yeah sorry I'm a pleb /g/ crossposter who comes here for non-retarded practical advice.

I'll definitely consider getting a hot air blower. But, doesn't heating the entire board up to solder melting temperature sound like a bad idea? I thought that avoiding that was the reason you shouldn't keep your iron pressed on the board for too long.
>>
>>1419498
i haven't read your posts but i desoldered some of my cherry switches on my keyboard just fine with an iron rather than a hot air gun. desoldering 3 or fewer pins by hopping the iron back and forth for a few seconds isn't difficult.
>>
File: IMG_20180706_000035.jpg (1.39 MB, 4128x3096)
1.39 MB
1.39 MB JPG
I found this small handheld toy in the back of a drawer. You have a choice of three buttons to press that make a loud SFX - an explosion, a gunshot, and a ray gun zap. Looking inside it's a really simple layout. There's a speaker and a small board with mostly nothing on it.

I'm only an electrician novice, but I have a question: is it possible to repurpose this toy to play a different SFX somehow? What is the process of doing this? The little chip is hidden underneath some black tar crap on the board. But what is the basic process of flashing a SFX to a new chip, or rewriting the existing chip? How is it done?

Any information is very appreciated.
>>
>>1419535
Can't really be done. Everything that matters is in that blob.
>>
File: IMG_20180705_235833.jpg (1.29 MB, 4128x3096)
1.29 MB
1.29 MB JPG
>>1419554
Presuming you can cleanly remove the blob, what then?
>>
>>1419569
Then nothing. You can't re-write them.
>>
This is more of a chemical question, but regarding batteries. I understand that the metals used are different charges, but do the electrons that leave the most negative out shell ever actually make it to the most positively charged electrolyte, Or do they just stop at the more positively charged core?
>>
>>1419498
it sounds like a great idea if you want to solder the whole board
for something like keyboard switches, a small hot air gun will be wonderful

>>1419569
you then get a microscope and reverse-engineer the chip under it ig?
more realistically, you cut the traces going to the chip and you substitute something else. the Nuvoton ChipCorder series of voice record/playback ICs would be appropriate, but afaict it's not trivial to make them play multiple samples as addressed by buttons. you wouldn't want to connect multiple modules to the same speaker, either. they could be damaged if more than one sound module plays simultaneously
>>
>>1419584
>you wouldn't want to connect multiple modules to the same speaker, either. they could be damaged if more than one sound module plays simultaneously
That's why you put a mixer IC between the modules and the speaker.
>>
anyone studying or already graduated fron a mechatronics career? wanted to get a technical degree but I wanna know more about work fields and the career itself
>>
>>1419593
then, having mixed line-levels instead of using the speaker amp onboard the ChipCorder ICs, you have to supply your own speaker amp, preferably bridged due to the low voltages involved, which is yet another IC or two, which then draws power continuously from the coin-cells (see image) unless you add extra circuitry to turn it on upon button press and off upon message-end signal from any of the ChipCorders, that circuitry itself taking some amount of standby power
not exactly trivial unless I missed noticing a member of the ChipCorder ilne or similar that will do multiple message playback based on button input
in other news, dem solder flags
>>
>>1419581
Never mind, I figured it out, all of the stuff I was reading failed to mention there was something between the anode and the electrolyte.
>>
>>1417865
The point of a grounding is to have a common "0" potential for your circuit. The electrical potential of your ground isn't necessarily 0V, but it is in relation to the rest of your circuit. This means that if you have more than one ground, they are most likely to be at different potentials which will lead to noise in your circuit at least. So you always want all of your components wired to one common ground.
>>
Brainlet here. I have questions about high-speed (~15 MHz) PCB design. I'm designing a board for my own homebrew 6502-based computer, and I want to know if there are simple things I should try to include/avoid in the PCB design to increase its chances of working at higher speeds (the CPU can handle up to 14 MHz).

The two important constraints are:
>1. I want to build the computer in pieces, with different sections on separate boards that all plug into a backplane. I am aware this is a bad idea and I want to do it anyway. This means some connections between components may be several inches in total.
>2. I want to keep all the PCBs down to 2 layers, because I'm a cheapskate. This means (I think) I don't have room for a ground plane.

I have no idea what I'm doing but an afternoon of furious googling suggests that the #1 thing I should try to fix is ensuring there's a good return path for each signal line. My plan for this currently is to interleave ground and signal pins on both the backplane connector and the PCB itself. Probably one ground line for every 2 signal lines (otherwise I'll run out of pins on the backplane connector). On the PCB I would run the ground lines alongside the signal lines until they get to the actual components, then the ground traces all go to the component's ground pin while the signals go to the appropriate pins.

Someone please tell me how retarded I am being. Is this the right thing to worry about, or would something else have better payoff / less effort? And would my proposed design actually help or is it just stupid?
>>
Are the circuit kits at places like Fry's worth picking up to practice a few things?
>>
>>1419724
download a simulator
>>
>>1419670
15MHz is not that fast, If you need to have really long lines and can't do differential signals, just try to have as good a ground plane as possible and remember plenty of decoooupling capacitors and the chips supply pins
>>
>>1419724
Go for the things you can't simulate, like the comically bad FM transmitter, or crystal radio.
>>
>>1419670
>how retarded I am being
only about 2/10
transceivers (e.g. 74xx245) and chip select/other bus logic near the card edge of each board are a good idea, as they shorten the stub lengths and capacitance seen by whatever's driving some signal on the backplane, and they reduce the fan-out that logic internal to the card has to drive
also, as other anon said, 15MHz isn't all that fast. ground/Vcc fill on both sides was enough for the old school designers, it should be enough for you, especially now that we have fat ceramic caps in the several uF range
also, you might want a good scope for this project, if you don't already have one
>>
Is there a name for circuits that don't use a pcb? It's kinda sexy.
>>
>>1419825
ugh. a disaster waiting to happen. at least put it in some potting.
>>
>>1419829
Clear resin would look nice. It's more art than anything.
>>
>>1419825
it's called dead bug prototyping but most people don't go near that far.
there's something perverse about using a prototyping technique to build a board designed for protoyping. maybe it was for hackaday cred or someshit.
>>
File: .jpg (91 KB, 800x432)
91 KB
91 KB JPG
>>1419835
>it's called dead bug prototyping
Thanks

>>1419829
>>1419832
Clear resin looks better than expected
>>
Could someone help me with creating an Electric Taser from the ground? Thanks.
>>
>>1419825
ugly style (yes, really)

>>1419841
nice try FBI
>>
>>1419841
Ok, first you will need some big capacitors, you can get them from any old CRT TV/monitor.
Before disassembling it, turn it on for a few minutes, that way you make sure the caps work.
Unplug it and open it, the capacitors are large cylindric things. You can remove them with a soldering iron (you know how to solder/desolder right?)
Before desoldering check the pins with your fingers, if they have a rough texture they are damaged, do not use them.
>>
>>1419825
Heard "air boarding" once or twice.
>>
File: Cordwoodcircuit.agr.jpg (528 KB, 1798x571)
528 KB
528 KB JPG
>>1419825

that reminds me of cordwood, which uses boards. or it used to, back when pic-related was all the rage.
>>
>>1419855
its all fun and games until you have to replace a cap
>>
>>1419849
yes i have some basic electronic skilles don't worry anon, what about that thing that real taser contain? btw great info thanks a lot.
>>
>>1419846
**KEK**
>>
>>1419868
You will need to fire the projectile (the projectile can be a couple metal nails attached together with some dielectric material). To fire it you use gunpowder. Smokeles gunpowder can be hard to get depending on where you are, so you can use black powder instead.
Mix one part of charcoal, one part of sulfur and six parts of potassium nitrate (it is sold as "stump remover")
Powder each component separately, then mix the powders. Then you pour it into a case and compact it hitting it with a hammer.
>>
>>1419670
>(the CPU can handle up to 14 MHz).

the 6502 comes in 3 speeds: 1, 2, or 3 Mhz. that means you need have zero concerns about PCB layout. you could even wire it up using wire-wrap, routing the wires willy-nilly, and it'll work just fine. i've done it with several 6809s at 2Mhz so i know.


the pic shows a wire-wrapped prototype Macintosh running a 68000 at around 10Mhz.
>>
>>1419825
Look up point to point shit as well. It's not really done for the memery like this though.
>>
>>1419880
>the 6502 comes in 3 speeds: 1, 2, or 3 Mhz
So he has actually a 65C02 or whatever.
Properly done wirewrap performs quite well. That said, a 65C02 running at 14MHz requires memories with a maximum read access time of 30ns, while a 10MHz 68k can manage with nearly 10 times slower memories.
Layout matters, but I'd expect the bus timings to be a bigger nuisance.
>>
>>1416580
I bought a bunch of electronic components. Sensors, transistors, basic supplies...
Suggest me things to do.

I also have an UNO and a NANO board.
>>
>>1419825
Short circuits.
>>
>>1419919
>Suggest me things to do.

Get rid of stuff you have no use for, and in the future, buy things that you need.
>>
>>1419919
this >>1419948, engineers call it the 5S
https://en.wikipedia.org/wiki/5S_(methodology)
>>
>>1416866
> higher price

It's just $150 m8. Nobody is asking him to buy an expensive R&S. Also

> Free decoding
> 1M fft
> Hackable to 200Mhz
> Bode Plot and WebServer out of the box.

It worth the price.
>>
>>1419957
>5S
That's for slaves. Makers can afford owning a junk box.
>>
File: cringe.png (1.02 MB, 989x719)
1.02 MB
1.02 MB PNG
>>1420004
>Makers can afford owning a junk box.
>>
File: 140623m.jpg (93 KB, 1067x577)
93 KB
93 KB JPG
>>1420004
>"at least I'm not a poorfag"
>>
File: IMG_0081.jpg (330 KB, 1604x1421)
330 KB
330 KB JPG
>>1420004
I love my junkbox, it's a source of inspiration.
>>
>>1419874
U are doing gods work anon, i love you.
>>
File: PCB.png (70 KB, 646x793)
70 KB
70 KB PNG
nixie tube anon back here again, I just finished this PCB for the supply from here http://www.vecoven.com/elec/hv_supply/hv.html
First time designing a switch mode supply, or any supply for that matter, and I did my best to follow some design guidelines. How'd I do? Is it serviceable?
>>
>>1419825
One of the golden rules the lab tech at college used to beat into use was to never ever under any circumstances depend on a soldered joint for structural strength. So when I see things like this my reaction is involuntary and violent.
>>
File: 1475513258126.jpg (342 KB, 650x867)
342 KB
342 KB JPG
>>1419850
I've used the term "sky-wiring", but usually in the context of a few flying leads out to something, vs. doing the whole assembly like that.
>>
>>1419919
>I have junk, what do?
Take it to the recycler.

>>1420234
>airwires
doesn't look finished to me :^) all topside ground plane? are you sending this off or making it at home?
anyway, think about using copper polygons instead of traces for power paths (+12V, that D2-D3-C5 node, ground?)
also consider putting some space around the high voltage paths, and not bringing the nodes of the multiplier so close together e.g. HV1 at D3 cathode, or the diodes in general (the leads could arc). personally, I'd probably do C-D-C-D-C-D all in a row, sort of rotating that whole section anticlockwise
does Q1 need a heatsink?
that schematic didn't specify resistor power rating. it's safe to assume he, being an old fart, meant 1/4W. 0805s are generally not rated for 1/4W. you might want to replace the 4x1 ohms with a single 0.25 ohm, 1W, in a 2512 package. also do some Ohm's Law calculation on the rest of the resistor ratings in power paths and make sure everything's good. if it's more that you just happen to have 0805s in stock, you could instead use 9x2.2 ohm to get close enough
why is the trace to the feedback pin so thicc? it's not carrying enough current to develop any real voltage across it. if it were thinner you could customize the TO-92 footprint to put more space between the base and E&C, and run the feedback trace between them
>>
>>1420270
I'm making it at home, and I already have that board on the bench and I'm literally midway through soldering it now. 0805 is all I have for SMD, and none of my thru hole sets have 0.25 ohm. I have some thru hole metal film 1/4W ones though, which of the 0805s would be important to replace with the thru holes? In the design guideline sheet from TI I was reading (http://www.ti.com/lit/an/snva021c/snva021c.pdf) it says to make the feedack trace relatively thick.
I'm just gonna finish soldering up the board as is and see if it works, and if it has any problems go back and make the changes you suggest and whip up a V2.
Oh and one last thing, the schematic specifies a 220uH inductor, but the closest relatively cheap and high current one I could find at my LES was 312uH. There's no issue with using a higher value inductor so long as it can handle the current, right?
>>
>>1420283
>>1420270
Alright, I soldered it all up and it works fine, and as far as I can tell, nothing is getting hot. Is there really any reason to redo the whole thing at this point?
Also, when I checked if stuff was getting hot I touched the backside while it was live like a dumbass, got a nice little shock.
>>
>>1420241
Do you get triggered by surface mount connectors that are ubiquitous in all electronics too?
>>
File: s-l300.jpg (17 KB, 300x287)
17 KB
17 KB JPG
What software do I need to use this atting85 programmer?
>>
File: power out detection.jpg (39 KB, 634x516)
39 KB
39 KB JPG
Hi, /ohm/.

Would this work for power-loss detection?

The "logic output" is a microcontroller pin, the led above is just simulating a load. The idea is that the OP-AMP will switch it's output state before the bigger cap can't power the microcontroller anymore, so it will know a power loss occurred and write what it needs to eeprom, and also go into a state where it's just idling (cuts power to the loads, etc) until the OP-AMP output is high again, then it can resume doing whatever it was doing.
>>
>>1420379
Arduino IDE and Platformio should work, BUT if you copy and paste the command line into Atmel Studio as a custom tool you can use it as well.

That thing has enough pins for a micro-sd card reader and it can be made to work as a HID device, meaning you can use it as a rubber ducky. It's pretty funny for pranks and hacking.
>>
>>1420283
>already have that board on the bench
kek, nice waiting for design review
>snva021c
huh. the same thickness as the other power connections would have been plenty. routing itself is good except for that oddball 90 degree turn
>which of the 0805s
in an abundance of caution, probably all of the 1 ohms, since they will be sharing current equally because of the equal resistance. the rest are probably okay, but you'll need to do the math on R1/R2/R3 (I = E*E/R).
otoh the 1 ohms might be just fine as they are. measure their temperature under full load, if you care enough
>312uH
that's fine, the inductor value is a minimum. as long as you aren't exceeding its saturation current (and you would know because your output would droop severely under load), you're fine
>>1420345
nah, if it works you done good. as for manufacturing, those acute angles potentially trap etchant and cut deeper into the traces, which is probably not important on thick-ish traces but can cut all the way through very thin ones
again, just for safety, you might put some heat-shrink or other insulating material between the diode leads that aren't directly connected, just to be safe. also consider the local humidity and maybe look up some clearance/creepage tables to see if you've put enough distance between them for the voltages desired

>>1420379
look up USBasp on the device side, avrdude on the host side

>>1420386
I think the most important thing is to make sure the micro has enough energy to complete its shutdown tasks, no matter what. those small resistors on the comparator aren't helping that at all. fix that first
standby power for micros is exceptionally low these days. can you just leave the micro always powered (thru diode-cap?) and connect the hard-switch only to the loads and to an input pin? then you can use a pin-change interrupt (which will likely be active during sleep) to initiate the shutdown process, *and* wake up the processor and have it restore operation
>>
Best way to go down from 24DC to 12DC?

I use a 7818 and then a 7812

I don't know much about electronics, I like messing around and on my way to learn
>>
>>1420409
the 7812 alone is fine, it should be able to take up to 35V on the input as long as it's attached to a heat sink
"best" way is usually a step-down switching converter, they're pretty easy to use these days if you can basic algebra. the MC34063 is widely used for all sorts of switching conversion because it works well enough, is cheap, available, and versatile, yet pretty easy to understand. search for the datasheet and dive in, and see how well you do
>don't know much
there are very good books in the OP. some of them can even be downloaded for free
>>
>>1420395
>no matter what. those small resistors on the comparator aren't helping that at all. fix that first
They're helping drain the PSU-caps faster, if anything
>>
File: IMG_4496[1].jpg (34 KB, 850x500)
34 KB
34 KB JPG
>>1417880
This is what I'm planning to use for my rpi-dash

https://mausberry-circuits.myshopify.com/collections/frontpage/products/3a-car-supply-switch-1

Not the cheapest thing but it solves enough problems to make it worth it to me.
>>
>>1420409
Depends on current draw. How many amps you pulling?
>>
>>1420262
That computer's wool coat is getting a bit thick, it needs shearing.
>>
>>1420619
32'hbaaaaaaa

>>1419825
sculpture, I suppose. I once read that old EM bombs were built in that same spirit, but with more elegant exploitation of electromagnetic principles
>>
>>1420518
not many, the circuit have a 555 and a 12v relay
>>1420463
thanks, I will try to experiment with the later
>>
>>1420639
you can easily determine where a linear regulator will work, and where you need a switching regulator. the power dissipated by the 7812 will be around (24V-12V)*50mA=0.6W, where 50mA is the upper end of what your relay might draw. check though, some relay coils draw more. at 0.6W you could probably live without a heatsink on a 7812 but if you want to be safe one of those shitty ebay heatsinks would do fine.
>>
How can I measure/process a data output from an eddy-current speedo?
I want to create a digital speedo for my car, it'd be easier but more expensive to do with GPS.
>>
>>1420730
Wouldn't it be easier to use an accelerometer and derivate its output?
>>
>>1420734
you mean integrate?
>>
>>1420734
That'd be a super simple option, but from what I've read, that method is super inaccurate unless you get some really nice sensors.
>>
>>1420736
>super simple option
you need to filter out the vibrations
>>
>>1420735
Yes that's what I meant, just a slip-up, I totally haven't forgotten everything I learned in school.
>>
>>1420739
Also find a way to account for accumulative error from numerical integration. Without "zeroing-out" at stopsigns it'd get less and less accurate.
>>
I'm not smart enough to do EE so I compensate with doing video game mods for people for money on the side as it makes me feel good that someone enjoys what I did for them and I get to work with soldering. I don't know what else to do with it. What are some other services/products I can do this with, I make about $800 a month doing this without trying.
>>
>>1420986
Is fixing TVs and stuff worthwhile nowadays? I tried to get into fixing laptops but it was never anything more than a broken DC jack or failed modular hardware and that's dumb as shit, you can't even get service manuals or schematics anymore. TVs are all modular too with daughter boards that you just swap out and I hate dealing with normies replacing phone screens and shit. I don't really know what else there is to really do actual repairs on.
>>
>>1420989
I wanted to get into BGA repairs, but you can't source the actual repalcement chips or they cost way too much to be profitable even buying $400 GPUs for 40 bucks. """reflow"""" and """reball""" as we know isn't an actual repair and it's just going to shit the bed soon anyway. I did try something where I would buy gaymer motherboards with bent CPU pins and replace the socket, but meh. What do you nerds do for pocket money? Sorry about the triple post I'm on the toilet.
>>
>>1420986
Friend I'm dumb as shit and I finished 4 years of EE.
>>
>>1420991
>>1420986
Dude if you know anything about computers just open up a repair shop. My mom fell for one of those indian-people-pretending-to-be-microsoft scams and when she went to the repair shop they wanted 200$ just to reset her pc to factory settings. You can charge old people whatever you want for a 20min job and they'll happily give you their money because computrons are way too difficult to understand apparently.
>>
>>1421006
Yeah I used to do that but like I said it's just really stupid easy and I honestly charge like $20/hr for the work I do now which can be up to running like 20-30 wires/sometimes directly to 100+ pin ICs etc, SMT rework blah blah. I own like $600 worth of equipment but I don't have any use for it outside of this because my clientele is poor or would rather use some 17 year old turd on eBay instead of me because he's cheaper. Trying to figure out some other hardware level repair service/product to switch to.
>>
>>1421012
I heard pinball/arcade repair makes a lot of money, but have no idea how to get into that scene. I can't just drop $400 on a broken cabinet and hope for the best, I only work part time at a grocery store and supplement with this stuff.
>>
File: 1505770133627.jpg (332 KB, 1000x667)
332 KB
332 KB JPG
>>1420649
a piece of copper clad might be enough heat sink for that application

>>1420767
>he doesn't rolling-stop
still no good for highway driving. what are the odds of just asking the ECU for speed info?

>>1420991
>six minutes
eat more fibre
>TVs
you might be able to make a few bob by picking up broken mid-range LCD TVs set out as rubbish and replacing what is probably just a blown transistor in a power supply, then turning around and selling them on craigslist or whatever. most parts should be easy to get from our guy ali
>>
File: vcf.png (41 KB, 1674x780)
41 KB
41 KB PNG
>>1418451
I was going to try to make a gilbert cell to use as a phase detector, but then I realized it's pretty much pointless or too much effort to do with discretes, and I discovered that a SA612 can sort of work, even though it's meant to have an oscillator circuit placed on one of the inputs, as opposed to another "input" signal from an op amp; but THEN i realized that a multiplier phase detector is pointless for my intended application because the output is 90 degrees out of phase with the inputs (holy run-on, batman!). My two inputs are (nominally) in-phase.
The idea is something i've been tossing around in my head for literally years without being able to come up with a realistic solution until now: an equalizer-type circuit where the filter bands follow the input frequency. Instead of having fixed bands spanning the entire audio spectrum, specific harmonics are adjusted. As you can imagine, simultaneously tuning 5+ voltage controlled bandpass filters to accurately follow a noisy signalfrom a guitar pickup over several octaves is a crapshoot, but a paper just published in 2016 shows a solution that might just fit inside of a large stompbox: https://www.worldscientific.com/doi/abs/10.1142/S0218126616500456
The bandpass filter for the fundamental frequency of the input is used INSIDE a phase-locked loop in place of the VCO, with the phase offset adjusting the control voltage. So instead of trying to independently tune an array of VCFs to ridiculous precision, the only requirement is to have each successive filter be tuned to a multiple of the loop filter's resonant frequency, which is easily accomplished with some precision capacitors. As for OTAs, LM13700s aren't prohibitively expensive.
requirement for through-hole components is just for ease of breadboard prototyping. I know this is a few lines of code with a microcontroller, but that's no fucking fun. Currently looking at phase-frequency detectors as the final puzzle piece, as they have 0 phase offset.
>>
Starting my first ever internship tomorrow lads, wish me luck.
>>
>>1421292
hopefully they don't abuse you too bad, but expect to be updating BOMs and putting together ugly documentation that the engineers don't want to deal with.
>>
I had this idea of controlling a power transistor with a variable base current. So I experimented by using a ULN2003 transistor array to vary the base current by driving a sufficient number of port pins high. This way the transistor's base could be driven by a sufficient current when its emitter current varies.
I'd like to measure the load current, and feed it into analog-digital converter of a microcontroller and it would then drive a correct number of port pins high. What kind of arrangement should I use for measuring the current?
>>
>>1421397
please just use an op amp and a pwm/dac. you can sense the load current with a small sense resistor but then you just need an op amp to amplify the voltage from that so your adc can read it anyway.
>>
>>1421397
>What kind of arrangement should I use for measuring the current?
hall effect current sensor
>>
What electrical components would someone getting into electronics need to stock up on?
>>
>>1421307
Apparently I'll be doing tests of their robot's SLAM algorithms, so hopefully it'll be more fun things than brainless work. I'm scared as fuck of not being prepared well enough on the technical side though.
>>
>>1421405
other than buying a shitty starter kit
resistors
capacitors (ceramic and electrolytic)
transistors
LEDs
operational amplifiers
diodes
voltage regulators
one or two arduinos (autists will tell you it is shit, but theyre just sperging)
>>
File: katen.png (96 KB, 297x237)
96 KB
96 KB PNG
Novice Picfag here. When I run this the GPIO 4 is not set, even though the instruction is being executed. send help picfaglords. All I want is to flash a led when I press something.
[code]#include "p12f675.inc"

FSR EQU 04 ; not used here
INDF EQU 00 ;not used
org 0x000
goto MAIN
org 0x004
goto EXINTR

MAIN:
bsf STATUS, RP0 ; goes to bank 1 to set TRISIO
bcf TRISIO, 4 ;sets gp4 as output
bcf STATUS, RP0 ; goes back to bank 0
clrf GPIO ;clears the io
bsf OPTION_REG, INTEDG ; sets to interrupt of positive change
bcf INTCON, INTF ; clears the interrupt flag
bsf INTCON, INTE ;enables external interrupt
bsf INTCON, 7 ;global interrupt set
clrf ANSEL ;sets stuff to digital io I think?

PT1: SLEEP
goto PT1 ;sleeps until interrupt

EXINTR:
btfss GPIO, GP4
goto BITSET ;jumos to setting the bit
bcf GPIO, GP4 ; clears bit
goto RETI ; exits interrupt
BITSET:
bsf GPIO, GP4
RETI:
bcf INTCON, INTF
RETFIE

end
[/code]
>>
>>1421415
>no code tags in /diy/
sorry famlam I'm a failure.

btw the GPIO port is the io port in this chip.
>>
>>1421416
>>no code tags in /diy/
>sorry famlam I'm a failure.
no la.
it is hirohito who has failed us.
>>
>>1421405
passives:
>a ton of assorted shitty resistors
>some power resistors
>assorted capacitors (both ceramic and electrolytic)
>pack of toroidal power inductors
>pack of small axial inductors

actives:
>2n3904 npn bjt transistors
>2n3906 pnp bjt transistors
>irf* power mosfets
>1n4148 small signal diodes
>1n4007 or eq power diodes
>whatever schottky diodes ebay has
>LEDs
>photodiodes/phototransistors
>zener diode assortment

integrated circuits:
>lm317 adj regulators
>assorted 78xx regulators
>555 timers
>uln200x drivers
>lm358 op amps
>lm393 comparators

suggested:
>pic microcontroller
>chinese pickit 3

all these can be found on ebay ridiculously cheap, like $1-2 for 10.
>>
>>1421400
I thought about using a small current sense resistor (maybe 1 Ω). But I'm not sure if it should be placed between +supply & emitter, collector & load or load & ground.
Time for another experiment, I guess
>>
>>1421428
load and ground
>>
>>1421415
are you not setting your configuration word or am i just blind? if you don't set it it'll default to FOSC2:0 = 111 which tries to use gp4 and 5 as an RC oscillator.
>>
>>1421434
I'm not. What's a configuration word? the datasheet never mentioned it.
>>
>>1421439
it's an extra program memory space that has some initialization options. it's on page 52 of your datasheet. i think you want 11000110010100.

example:
INCLUDE <P16F1454.INC>
__CONFIG CONFIG1 B'00100110100100'
__CONFIG CONFIG2 B'01111111001111'
>>
>>1421443
thanks my man. I somehow missed that
>>
>>1421523
>>1421443
The problem was that I forgot to turn off the analog comparator and set the pins to digital I/O. It's fixed now. That is done setting the ANSEL and CMOM registers

#include "p12f675.inc"
; CONFIG
; __config 0x3FDC
__CONFIG _FOSC_INTRCIO & _WDTE_ON & _PWRTE_OFF & _MCLRE_OFF & _BOREN_ON & _CP_OFF & _CPD_OFF
FSR EQU 04 ; not used here
INDF EQU 00 ;not used
org 0x000
goto MAIN

org 0x004
goto EXINTR

MAIN:
movlw b'111'
movwf CMCON ;turns off analog comparator
bsf STATUS, RP0 ;goes to other bank
clrf ANSEL ;sets shit to digital IO
clrf TRISIO ;sets all GPx to output
bsf TRISIO,2
bcf STATUS, RP0 ;go back to bank 0
clrf GPIO ;sets GPIO
;now to set up the interruption;
bsf OPTION_REG, INTEDG ; sets to interrupt of positive change
bcf INTCON, INTF ; clears the interrupt flag
bsf INTCON, INTE ;enables external interrupt
bsf INTCON, 7 ;global interrupt set
REST:sleep
goto REST
EXINTR:
btfsc GPIO, GP4 ;is the bit 0? if yes skip clearing bit;
bcf GPIO, GP4 ;clears bit
btfss GPIO, GP4 ;is the bit 1? is yes skips setting the bit;
bsf GPIO, GP4
bcf INTCON, INTF
RETFIE
END
>>
File: 1531086179282.jpg (594 KB, 2070x1743)
594 KB
594 KB JPG
>tfw ruined pads
There isn't a way to salvage them, is there? When I tried to solder the LED the solder literally didn't stick to the PCB at all no matter how I tried, it would just collect as a glob on the LED leg and the tip of my iron. I assume this is a lost cause?
>>
File: SD Card Module.jpg (78 KB, 800x800)
78 KB
78 KB JPG
Does anyone have a good schematic for a SD card reader with SPI? I have seen a few but i'm unsure how well they work.
>>
>>1421614
you could try following the traces and running some wires from the other side of the traces to the the LED
>>
>>1421620
dude, that board has the names of the pins written on the other side
connect them to those with the same name on the MCU, thats it
CS is slave select, you can connect it to ground if you dont use any other SPI devices on that same MCU
>>
File: SD Card Circuit.gif (5 KB, 520x330)
5 KB
5 KB GIF
>>1421624
I was looking more for a circuit of the module similar to pic related to later put on another circuitboard.
I have seen many different looking circuits and thougth i ask you guys if any one had a tested circuit or knew of a well working design.
>>
Does anyone here habe pdfs or the like of Widlar's books?
>>
>>1421629
The only thing the board does is interface between 5v and 3v3 logic. That diagram seems right, except that the board has some more caps and resistors, and also it looks like it doesnt have MISO connected directly like that, looks like it goes though the 4th buffer gate
>>
>>1416580
does anyone know anything about USB-PD or QC 3.0? I'm looking for a small IC or module that can trigger the different voltages of QC, think like those cheap Ebay voltage testers. I want to stash it in my laptop so that I can charge it with a macbook charger (QC compatible as it "fast charges" my note 8)

I want kinda a unified system that just uses one charger for all my shit
>>
>>1421646
>macbook
>>
>>1421648
I actually own 8 thinkpads and an elitebook, the macbook is my girlfriends.

why would I want to convert my laptop to use a macbook charger if it was already a macbook.

fucking 4chan man. smartest people in the world mixed with people like you
>>
>>1421649
inb4

>redit spacing

this is the first 4chan post I've made after several weeks of writing emails with normal spacing
>>
>>1421649
>having a crapple buying gf
even a tranny would be better
>>
>>1421652
>unironically thinking apple is bad
apart from their repair policy they are okay man, on the high side of price but you pay for the form.

this is coming from someone who owns
z61p
t400
t400
t500
w500
x201s
t420s
x230t

elitebook 8760w

they work fine and look good too.
get over it
>>
>>1421654
>get over it
no
apple = shit

>look good too
everything made by apple looks like shit
you should be ashamed
>>
>>1419836
single op amp headphone amplifiers are bad enough. Then they sealed some huge electrolytic caps inside solid resin with no way for them to vent if they fail.
>>
>>1421663
okay sweetie, but let's get back on topic.
I'm thinking of ripping apart one of those test meters but my concern is getting them to trigger the 20v signal each time
>>
>>1421672
try this
http://blog.deconinck.info/post/2017/08/09/Turning-a-Quick-Charge-3.0-charger-into-a-variable-voltage-power-supply
you still gay
>>
>>1416872
>>1416891
all you need is your penis and you can probe any circuit
>>
>>1421424
>power resistors
What do you even do with these?
>>
>>1421652
>a tranny would be better
>implying

>>1421906
you resist power, of course. common uses are measuring current according to I=E/R, providing ballast for a shunt regulator especially with a wide Vin-Vout spread, and enforcing current sharing between parallel transistors
>>
File: AD594-595-fbl.png (104 KB, 900x1145)
104 KB
104 KB PNG
For the life of me, I can't get this stupid circuit to cooperate with me, when my motivation to build it was for the fact that it was simple. Thought maybe someone here could give me their input.
So what I'm trying to do is use a thermocouple to measure the temperature of an oven, and output it to my laptop via an 8051 dev board (that I made in school ages ago). The circuit I'm using to measure said temperature is this AD595, whose output I connect to a MCP3008 so that my 8051 can read the temperatures.
When I hook everything up and output the temperature to my laptop, I get readings that are 5-10 degrees higher than what the room temperature is. When I touch the thermocouple probe to warm it up, the temperature starts going down back to the actual room temperature. As soon as I let go of the thermocouple, the temperature immediately spikes up to something like 40C and slowly goes back down to being around 5-10 degrees above room temperature.
1) I know for a fact that the 8051 is not the problem. It works for every other application that I use it for, so I know that the circuit I'm plugging to it is the issue.
2) I isolated the AD595 from the MCP3008 and tried measuring the output with a multimeter. To my surprise, the voltage reading I was getting was exactly room temperature. I was able to rise the temperature by warming the probe with my finger. Clearly, the AD595 is not the problem.
3) I checked to make sure all the pins on the MCP3008 are connected to their respective pins on the 8051. No issues there.
4) I have tried switching the MCP3008 with 3 other fresh MCP3008 ICs. Still no dice. The MCP3008 is not the issue.
5) Just by accident, I left the negative probe of my multimeter connected to the ground of my circuit while running my program to read temperatures. This apparently solved the issue because the temperature readings were being spat out perfectly, and would change based on whether I'd touch the probe or not.
>>
File: Circuit_Diagram_AD595.jpg (169 KB, 608x1080)
169 KB
169 KB JPG
>>1421929
Anyway, I figure it's some sort of grounding issue that's causing these erroneous readings. But that's where the truck stops, I'm not sure how to go about solving this issue.
This was all done on a perf board since I know thermocouples can be a little sensitive to the extra metal in breadboards.
I should also mention that I did have this working perfectly fine at one point. I left it alone for about a week, and when I came back to it (didn't modify it in any way), it wasn't reading temperatures properly.
It's just frustrating because this should in theory be the easy way out of messing with OP Amps.
>>
>>1421929
>If the thermocouple is not remotely grounded, then the dotted line connections in Figures 1 and 2 are recommended. A resistor may be needed in this connection to assure that common-mode voltages induced in the thermocouple loop are not converted to normal mode.
if a resistor doesn't work, what happens when you insert a voltage follower on the output of the AD595? pin 8 looks like a great place to couple noise back into the system
>>
>>1421929
Makes me think your sw is reading a wrong (adjacent) ADC channel. Or, if it's not specifically that, at least you have something floating.
>>
>>1421663
are you the same pajeet that was of /dpt/ on /g/butthurt about people having more money than you to actually have a choice of buying overpriced bad products? The problem isn't only apple, it's you.
>>
>>1422017
>are you the same pajeet
no, im not a pajeet

>defending crapple
maybe you are pajeet
>>
I bought this https://www.banggood.com/TK2050-50W-50W-Audio-Amplifier-Board-T-Class-Dual-Channel-HIFI-Stereo-p-1070906.html

It has holes for a pot for volume control

How do I figure out what size pot to use?
>>
>>1422053
https://www.manualslib.com/manual/734008/Sure-Electronics-Tk-2050.html?page=10#manual
and remember you use log pots for audio volume
>>
>>1422057
Thanks bromeslice
>>
Can someone explain to me what is meant with IC1A, IC1B, IC1C, IC1D and IC1P? They are a part of the 74ABT125PW IC somehow..?
>>
>>1422215
>IC1A, IC1B, IC1C, IC1D
the 4 buffer gates
>IC1P
power
>>
>>1422217
Thanks
>>
File: Untitled-2 - Copy.jpg (818 KB, 4078x2544)
818 KB
818 KB JPG
I have a Mesa/Boogie V-Twin guitar pedal that I want to fix. It works fine when you first plug it in, but after about 10 minutes or so the volume drops so low you can barely hear it. It seems to sound the same though, just significantly lower, so if the gain is set high it's still very distorted as it should be.
The clean channel keeps working as it should.

I found this schematic and marked some stuff on it that may or may not be correct as I don't really know anything about this shit. My current idea is that one of the TL072 op-amps is dead.
Does this seem reasonable or could it be something else?
>>
>>1422248

op-amps hardly ever go bad. tubes, on the other hand, are born with one foot in the grave.
>>
>>1422252
There's nothing wrong with the tubes. I tried swapping them, and it didn't fix shit.
>>
File: ircodes8.png (3 KB, 480x208)
3 KB
3 KB PNG
How do I generate a modulated TV remote signal with my attiny85?
It needs to be precisely 38kHz, but only turn on at specific times.
I was thinking about using two timers, one running at 38kHz and other triggering periodically every 600us, so a 1 would be two ON periods of 600us followed by one OFF period of the same time, and a 0 would be one period ON and one period OFF.

Would combining two timers be a sensible idea? Is there a better way to send a 38kHz signal inside a longer one?

Pic related.
>>
File: Screenshot (3).png (478 KB, 1600x900)
478 KB
478 KB PNG
complete beginner. hows my selection for a starting point?
>>
>>1422277
if you can generate the 38khz carrier either externally or with a separate on chip peripheral (like pwm?) and then combine them in hardware it will be easier. to write software to do it is of course possible but you might need to figure out the timing. i mean at 8mhz on attiny you have what 200 instruction cycles to sort yourself out? should be easy enough.
>>
>>1416580
What is a good multimeter
I've only used the ones from Horror Fraught and I want to upgrade but I don't want to break the bank
>>
>>1422282
link to that kit
>>
>>1422277
i don't know (((atmels))) but you can use the timer peripheral to create a 38kHz pwm derived from the internal oscillator, which is probably accurate enough since the ir receiver is just a shitty bandpass filter. control the pwm frequency using OCRnA and modulate the signal in software using the COM0n0 bit.
>>
>>1422298
https://www.amazon.com/gp/product/B01ERPEMAC/ref=ox_sc_act_title_1?smid=A2WWHQ25ENKVJ1&psc=1
do i need an arduino to use this stuff
>>
>>1422317
you dont *need* the arduino, but it will let you do a lot more interesting things

with just that kit you can light leds, build logic gates, use a shift register (for something?)
they should have included at least a 555, how would you drive those buzzers otherwise?
>>
how to attach stuff to a dc motor shaft? shit like a drill chuck.
>>
>>1422371
that's more mechanics than electronics
simply, you need a chuck that can fit on that shaft, or use an adapter, but that sucks
>>
How would I go about determining the output impedance of a voltage multiplier?
>>
>>1422384
You would measure it.
>>
>>1422398
I take it then there's no method (or at least no easy method) for calculating it?
>>
>>1422384
>output tension without load
>current with load
>division
there you have it
>>
>>1422402
>no easy method
Correct. Study the relevant literature re charge pump.
>>
File: file.png (177 KB, 600x600)
177 KB
177 KB PNG
>>1422371
shaft coupler. anything under load will need something more robust than the ebay special though.
>>
>>1422411
>>1422413
Well the input to the voltage multiplier should be a square wave with a duty cycle between 5 and 60% at 180Vpk coming off an SMPS. I'll probably use ultrafast rectifier diodes like the UF4007 and I simply want to double the voltage for 360VDC out. The maximum output current I'd like to be able to achieve is 100mA but I'd settle for 50mA. To achieve 360VDC 100mA I need an output impedance of 3.6kΩ or less. I simply was looking for a way to figure out what size capacitors I'd need before I built the prototype.
>>
>>1422419
a much better answer than "study the literature" is literally just to throw it in a simulator, see what you need, and then tack on some margin.
>>
>>1422419
>To achieve 360VDC 100mA I need
36W. Good luck.
>>
>>1422419
>To achieve 360VDC 100mA I need an output impedance of 3.6kΩ
thats for your load
the output impedance of a supply circuit determines how much the voltage drops when you connect the load
>>
>>1422437
Ah right my bad, I really need 360Ω max preferably less so my voltage doesn't get divided down too much.

>>1422429
I'm just over provisioning since I don't know exactly how much current my load will pull. Looking at a random schematic for a similar design to mine the diff amps this guy is using to drive the deflection plates have a bias current around 300uA so it's possible I don't need more than a few mA on the + and - 360V rails.
>>
>>1422442
the impedance of the circuit depends on the frequency
>>
>>1422447
Input to the multiplier is 50kHz 180Vpk square wave with a duty cycle between 5 and 60%. Output is 360VDC. There is no frequency on the output if you don't count the ripple.
>>
>>1422451
If I understand the circuit, the current flows out of the caps, so there should be a rather low impedance, don't worry about it, too small to care
>>
File: board schematic.png (117 KB, 2338x1653)
117 KB
117 KB PNG
Is there anything wrong with this schematic? I'm having trouble getting the board for it to work, the tubes attached to it work intermittently, and tend to stay on the same numbers or flicker randomly when they do turn on, but I can't find any problems with connections anywhere.
>>
>>1422460
>If I understand the circuit,

you dont. since it's a high voltage circuit, the caps are either very small or very expensive. since everyone opts for cheap, the caps are small and the output impedance tends to be ridiculously high. just pulling 1mA is gonna dip the voltage significantly.
>>
Anyone willing to take a look at this? The tubes are not the problem. I would get the scope from the attic but it would be nice to know what to check since I don't have a tone generator.
>>1422248
>>
>>1422567
I'm looking at ceramic, mica, or metal film capacitors in the range of 10 to 100nF with a minimum voltage rating of 400V, preferably 600V+, so yeah, nothing that's gonna hold a large reservoir of charge. Fortunately I think I was way off with that 100mA max current estimate. Seems like 3LO1I needs very little current to drive all the anodes, cathode, grids, and deflection plates, disregarding the heater which is run of a separate 6.3V supply not derived from the B+ rail. As long as the output of the voltage multiplier can source a few mA of current I should be good.
>>
>try and find literally any resource at all on EE/CE
>if it's a video it's a well nigh unintelligible pajeet
>if it's a written article it's a well nigh unintelligible pajeet
literally what the fuck
they can barely do IT work, why are they so prevalent in EE
>>
>>1422588
Population of India: 1.324 billion
>>
>>1422588

also, if you can communicate well in English, then you can charge top price for your work. it would be insulting to work for pennies on Jewtube, like incomprehensible pajeet does.
>>
>>1422573
my first guess is secondary emission or power supply out of spec. check heater voltage, +12, -12, 5V supply, and also the 12V AC input (you can check voltages at the VI pins of the 7x12s as an estimate of peak voltage in their respective direction)
check V+ (careful, might be a few hundred volts)
check dc voltages on V2A and V2B (all pins) with no signal applied
check resistance across R38
check voltage on gates of Q1, Q8 and Q6 with respect to ground
record all these measurements on cold start (with the chosen gain mode activated) AND after it starts to fail, and you might see a pattern
>>
Did we hear back from "20W voltage regulator" anon yet?
>>
File: 1496166031810.jpg (66 KB, 639x629)
66 KB
66 KB JPG
>>1422294
anyone
>>
>>1422743
Have a look at the EEVblog multimeter spreadsheet. The Angeng 8008 might have what you're looking for, pick it up on ali or ebay for pretty cheap.
>>
Anyone knows of good schematics for a spot welder? everything I find online is just "lol dood just get this capacitor and this microwave and lmao"
>>
what are the chances of they exploding on my face?
https://pt.aliexpress.com/item/New-Arrival-1-PC-Black-Farad-Capacitor-2-7V-10F-10MM-Diameter-26MM-Length-Super-Capacitor/32798790034.html
>>
>>1422588
they're cheap
university-level resources tend to be alright. you can get a lot from the slide decks even without the lecture that goes with them

>>1422779
a spot welder is a basic exercise in getting a few volts at many amps to a couple of electrodes. the magic, as it were, is in the construction and materials selection. that kid with the 600F capacitor or whatever had a decent light-duty spot welder. that guy who rewound the microwave transformers had a decent, more heavy-duty spot welder. don't like their designs? learn from their experiments and use what you know to make your own thing that dumps hella current

>>1422783
high, if you abuse them
reasonably low, if you operate them within their stated specs
>>
>>1422790
it's not that I don't like their designs, it's that I don't understand what the fuck they are doing, I don't have much experience with electronics, so I don't know what paremeters to aim for and as consequence what parts I should use etc. Most sources I got barely show their power circuitry. (One was a guy literally connecting a pushbutton to a car battery)
>>
Does anybody here with experience with kicad know how to get some kind of marker working to show the location of an unconnected track. I get 3 unconnected tracks in the drc but they are increadebly hard to find if I have no idea where to find them.
>>
>>1422964
In 5.0 the DRC window has a "List Unconnected" button, and you can click on an item in the list to center the display on it.
>>
Someoen make a new thread please so i can show off my semi dodgy diy lamp
>>
File: V-Twin.jpg (1.78 MB, 4160x3120)
1.78 MB
1.78 MB JPG
>>1422629
I appreciate you taking the time to actually look at it. I opened it up, and it works fine as long as you keep the lid off. Put it back on and it craps out after 5-10 minutes. Take it off again and it goes back to normal after a while. The tubes aren't really enclosed.
Heater voltage and V+ would be the same for both tubes, no? AC input is a bit hot at 14V but that shouldn't really matter either..?
>>
>>1422282
kek, i got an EGLOO resistor kit for the box. But for the price, check ebay. You can get a ton more for half the price
>>
>>1422322
nowadays kids aren't interested in descrete digital logic. Its all "Arduino" now.

Atmel, you helped kill hobby electronics
>>
>>1423036
Even if he was interested in discrete digital logic, there's not a lot you can do with five transistors and a shift register
>>
>>1423046
but there is plenty to learn. So much more interesting than seeing a bar go along the screen
>>
>>1422993
Now you need to reproduce the error condition with the lid off so you can measure and log (!) data/observations to narrow down and find the fault. Does 'hair dryer' ring a bell? Or a small halogen lamp at close distance (my favourite for such cases). You have the circuit diagram and you have a multimeter, optimal conditions.
>>
>>1422993
>supply voltage 16% too high
that could matter. every unregulated voltage in the system is based on that, including the heaters and V+. when the plate of a triode gets too hot, whether by heat radiated by the cathode or by being hit by too much current, it starts emitting/repelling electrons much like the cathode, and down goes your gain
I'd try to fix that supply voltage first, whether with a power resistor in-line or a new transformer, or, if you really don't mind looking dorky with a shitty workaround, add a fan
>Heater voltage and V+
yep, heaters are same for both, but plate voltages will differ depending on operating points for each stage

>>1423036
>discrete digital logic
I see literally nothing wrong with this. probably better if kids these days learn Verilog
>atmel
don't blame Atmel, blame that MIT chick for creating a piece of shit
>killing hobby electronics
>implying 74xx crap is the heart of hobby electronics
>implying the hobby shouldn't follow the industry
yeah nah
>>
>>1423036
in real stuff you usually use both? Sometimes it's a billion times cheaper to throw a few bjts (and more robust too) than putting another pic
>>
So I got my paws on 12 or so phone chargers, I plan on connecting them in series, stuff in a box and split the output to make a bench PSU. Is that a bad idea?
>>
>>1423151
this is true... and there are still a very few 74xx-series logic ICs that are useful, not counting the always-useful small-package 1G/2G series

>>1423154
terrible idea. they won't have the voltages you need when you need them

>>1422974
>angsty teen mode
fiiiine

NEW THREAD
>>1423156
>>1423156
>>1423156
NEW THREAD
>>
>>1422792
its fairly simple, all you need is a way to turn on a low voltage, very high current supply for a short, measured amount of time
>>
>>1423141
I can just attach it to the lid by the pots and turn it upside down. Maybe block the sides with whatever. Should work fine, especially since it's upside down.
>>1423148
>including the heaters and V+
Only the heaters and V+. V+ probably isn't a problem with the voltages they're running at, but heaters could be from what I've gathered (in terms of tube life at least). I'll probably check the voltages at the tube sockets tomorrow.
>add a fan
The idea did cross my mind, but it's a bit too valuable to niggerrig.
>>
>>1423267
>niggerrig
african re-engineer
>>
For whatever reason, the circuit is working now. I have been messing with it in terms of checking it with a multimeter, as well as disconnecting ground/vcc and plugging them back in. Might be worth noting that the 8051 dev board is connected to a breadboard, and I'm using the vcc/gnd rails to power my circuit. This lack of consistency bothers me, I don't want to shove this circuit into an enclosure and have it only work sometimes.

>>1421951
I saw that bit in the datasheet, but how large of a resistor would I need? I think when I was messing with the circuit on a breadboard, I tried using a 10K ohm resistor, but that didn't do anything. Granted, it wasn't working at all because I was testing the circuit on a breadboard, which wasn't good for the thermocouple.
I don't think I have any OP amps laying around that I could make a voltage follower with. Would that really do anything though?

>>1421954
Software was written by my brother. Issue is, sometimes it works, and sometimes it doesn't.
My circuit is laid out exactly as I have it in >>1421933
>>
>>1423267
you should try to measure the voltages with the tubes in place
also note that the plates are just a couple of millimeters from the cathodes. if they are running too hot the plates will heat up more than they should

>>1423323
you'll have to try it. off the top of my head I'd say 10k would be enough
>Would that really do anything though?
if you're getting noise back from the ADC, it could help
>lack of consistency
you might put your cell phone by the device under test and watch for interference
>>
File: 1491800192283.gif (780 KB, 400x250)
780 KB
780 KB GIF
If I have a solar panel connected to a 12 volt battery can I connect a small 35 watt fan to it?

Wouldn't it short out the fan because the battery has a lot of power in it?

explain for a retard pls
>>
>>1423341
>>1423148
>>1423141
Update here >>1423560
>>
>>1423473
>explain for a retard pls
Why doesn't the 12v battery in a car short out the 1 watt courtesy light when you open the door?




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.