[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y ] [Search | | Home]

/ohm/ - electronics thread, wikity wikity wack edition

This is a blue board which means that it's for everybody (Safe For Work content only). If you see any adult content, please report it.

Thread replies: 330
Thread images: 73

File: 1490667328763.jpg (82KB, 680x1024px) Image search: [iqdb] [SauceNao] [Google]
1490667328763.jpg
82KB, 680x1024px
bump limit reached on old thread >>1226748

pastebin.com/9UgLjyND (stale)
https://www.wiki.printf.pl/index.php?title=Pasta (fresh)

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

>What books are there?

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

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

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

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

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

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

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

>What software should I use to layout boards?
Circuit Wizard
ExpressPCB
EAGLE
KiCad
>>
>>1230603
Damnit, the only reason that image was posted in the last thread was because it was someone's OP earlier in the year/last year.
>>
>>1230606
I do seem to have a talent for pissing everyone off with the photos I post. Maybe my username on the new wiki should be badpixanon.
Fine, then. You don't like the holding the wrong end of the soldering iron series? Post something better and I'll delet.
>>
I am looking for some advice. I have done all the research I can and bought several sets of plans but they are all extremely dated i.e. calling for parts of a Model T car.

I am trying to build a sideshow Electric Chair. These are chairs or stools you can sit or stand on, turn on, and then when turned on light lightbulbs in your hands and lightly shock people. All I can gather is that I need a power source of some sort and a copper plate to make contact with. If anyone has any ideas it would be greatly appreciated. I can provide links to videos if anyone would like to see one.
>>
>>1230608
I don't dislike the image itself, I'm actually just amused by other people getting pissed off. I just think we need a better selection of good, shitty, all sorts of images. Maybe the wiki could be that.
>>
>>1230610
Those are based on Tesla coils, if I'm not mistaken. Some care is required.

>>1230611
Fair enough, and agreed.
>>
>>1230598
Probes might be worth most of $100 without the scope. Noice.

>>1230605
Oh, okay. So you can't easily get to the low-power side to do some more efficient filtering, without using a software EQ or something like that. Good enough m8.
>>
File: 1497060655767.jpg (76KB, 1000x1000px) Image search: [iqdb] [SauceNao] [Google]
1497060655767.jpg
76KB, 1000x1000px
what's a good soldering iron /ohm/? I have that shitty $10 one that doesn't even have variable temp so I'd like to upgrade.
I've seen these 24V irons being recommended all over the place... are they a meme? For $60, they're fucking expensive.
>>
>>1230614
That's what I was thinking, I know very little about oscilloscopes but I do know good probes are fuckoff expensive.
I love that it comes with the manuals too, I get a boner for big chunky manuals of any kind.
>>
>>1230618
Get a knockoff Hakko station from Yihua or something, for a quarter of the price of the real thing they're excellent and will probably do you for years, if not more.
>>
>>1230613
I have spoken with an old timer who said he used a neon tester or something along those lines? A motor generator? He didn't provide much more info than that. The older versions (way back) used Tesla coils or violet wands. Any idea how to implement a neon tester or motor generator? What voltage? To be honest I am completely foreign to electrical stuff and unfortunately everyone in my line of work keeps just about everything as an "industry secret".
>>
>>1230620
The thing is tho... these TS100 open-source irons are performing better than Weller and Hakkos.
>>
>>1230662
>open-source
Shit, really? Now I want to buy the parts and make one out of wood. I guess it's not hard to design one of these from scratch, so I guess the main thing to do is find the ideal element + tip combo. I could probably go full max-wattage and no thermocouple and measure the temperature through resistivity alone if I made enough measurements, but I'd need to calibrate the iron(s) pretty precisely with a bunch of testing for each element, so it's probably not worth trying to mass-produce them. 1kW iron when.
>>
>>1230598
4 channel, 100 MHz is not bad for $100. Though sampling rate vs BW on early digital scopes can be kinda iffy sometimes.
>>
>>1230677
watch...

this dude tests it out against Weller
https://www.youtube.com/watch?v=HgrB5P-rDLw

my favorite drone dude analyzes it

https://www.youtube.com/watch?v=t621xQc-xEQ
>>
File: 1497254410826.jpg (368KB, 800x800px) Image search: [iqdb] [SauceNao] [Google]
1497254410826.jpg
368KB, 800x800px
>>1230662
I've got a Xytronics XY-168 made in the early 1990s or before, passed down to me by my mentor, that's falling apart but still working and mostly holding temperature. How long do TS100 tips last? I would really rather not be screwed in case the burger govt does something stupid.
Pic related might be the next OP image.
>>
>>1230682
>>1230603
>burning_fingers.jpg
>>
>>1230697
No problem, it's not turned on.
>>
>>1230682
Assuming those elements run off 12V or so, there'd be nothing wrong with putting 150V through them for short amounts of time, right? Short enough so that it doesn't overheat, I mean. Or would the element's insulation not be good enough to trust at those voltages?
>>
>>1230699
Be the first to find out.
>>
>>1230618
http://s.aliexpress.com/B7Vr2ym6
This, or another kit for T12 tips
>>
>>1230699
I think you might blow up a few components or two. Why do you even want to anyway?
>>
File: ts100_4x.png (286KB, 3364x2380px) Image search: [iqdb] [SauceNao] [Google]
ts100_4x.png
286KB, 3364x2380px
>>1230699
They run off 24V and if you're only using the element then yes, but if you are using the TS100 then there's a DC/DC converter on the input (RT7272B) and 150V would fry it in moments (specified for 36V input max, so I guess putting a little over 24V into the iron wouldn't be a problem).
>>
I've never really been around on these threads before, but I might check them out now and again.

I've been worrying that I've been spending too much time on hardware projects and I would want to try and do something embedded wise and I have been thinking of trying to make a USB audio interface. Would anyone recommend any development boards to try and build this up, concept wise?
>>
>>1230723
Those cheap STM32 boards from China have USB support, but when I looked into the provided examples USB audio was the only one missing. From what I've heard STM32F4 have some nice USB audio options, look here:
http://www.tjaekel.com/DiscoveryUSB/index.html
>>
>>1230727
These are a development board all by themselves, right? It does look interesting, and for $27 AUD from mouser, it doesn't sound too bad.

Briefly looking over the microcontroller itself, it seems to have 3 ADCs and 2 DACs, but its confusing what the board itself has pinouts for.
>>
File: TS100-caution-temp.png (35KB, 906x314px) Image search: [iqdb] [SauceNao] [Google]
TS100-caution-temp.png
35KB, 906x314px
>>1230682
>How long do TS100 tips last?
Almost forever, because you will rarely use it. Why? Flawed thermal design. It burns your fingers after half an hour or so, even at 12V=17W. /ohm/'s running gag has made it into real life. The TS100 is not a tool, it's a toy. But you can do firmware updates all day long..
>>
>>1230732
I always thought that the TS100 looked fairly nice, but it just seemed like too much dicking around with cables and you don't get a stand with it, it didn't seem like a useful tool in my eyes.
>>
>>1230718
>>1230714
I was referring just to the element, terminals A and B in the pic. Looks like the element is connected straight from V_DC, and is not regulated down to some value, meaning the thing will heat up significantly quicker if plugged into 24V compared to 12V. The modifications to get it to run off 150V/rectified US mains wouldn't be too difficult, but you'd probably fry the element because of the lag between the element itself heating up and that heat reaching the temperature sensor, though looking at the diagram I can't see anything that looks like a thermocouple, and the tips only have two terminals on them. So then can we assume that the temperature is measured by the opamp measuring the voltage across the element that has 3.9V across it through the 39kΩ resistor? The opamp feeds into what I assume to be an analogue input of the IC, and the voltage regulators both appear to be switching regulators, so the actual modification required to turn this into a 1kW iron would actually be fairly minimal.

Can anyone with a better sense of this schematic confirm? Is the second of the two opposing MOSFETs between Vin and A there so that when only powered by USB your programmer can tell you what the baseline sense current through the element is?
>>
>>1230723
Greetings and welcome.
How about Teensy 2.0? AVRs have the advantage of the highly recommended LUFA framework http://www.fourwalledcubicle.com/LUFA.php which includes USB audio in/out among its examples. It might be easy to change the existing dual-channel PWM output for a serial DAC. Not sure about I2S due to the great likelihood of jitter.
>>
>>1230737
Hmm, I'm not sure. I feel as if that that stm32 f4 discovery board would be better suited to the task, however I feel as if that the Teensy board would actually have better support.

I haven't really done much embedded stuff apart unless it was related to studies, so I honestly don't know.
>>
The OP image got me thinking, /ohm/: when did you first burn yourself on some component? It feels more like a constant of electronics than anything else.

>Why isn't this relay switching?
>My 9v must be goosed
>Accidently touch a mis-wired transistor
>!!!
>>
>>1230766
On a component? It wasn't actually my component, I was going my physics department's stage 2 "basic electrical measurements" lab and I was testing the accuracy of the multimeter just to begin with to make sure the fuse hadn't gone (also first time touching a Fluke, breddy gud) and I calculated the current I'd be putting through a resistor to make sure it was below the multimeter's current limit. For 12V and the so-many hundred mA value, I calculated it would be safe to use a resistor of somewhere between 50Ω and 100Ω. I hooked up the component and made my measurement, taking my time while a wisp of smoke slowly exited the resistor, at which point I yanked it off, calculated the power dissipated though the resistor, and realised it was multiple watts. It wasn't a burn bad enough to leave a blister, but I definitely noticed it. Apart from that my only burns have been due to the heat from the soldering iron, either directly or accumulated on a component or PCB, and also from my high-school's gas-discharge-tube power supply which I was messing around with plugged into my MOT's secondary. I was wearing cycle gloves at the time to prevent shocks, and I'd had a few too many scrapes beforehand, leaving one glove slightly perforated, making it less than ideal for insulation. Surprisingly the shock was just a slight belt and wasn't deadly-feeling at all, so I assume it wasn't running at mains frequency but rather somewhere in the 100kHz+ range.

I have picked up a hot soldering iron like OP's pic at least once, and a non-temperature regulated one too so chances are it was 300°+.
>>
>>1230735
>can't see anything that looks like a thermocouple
because there is none. The TMP36 is specified for a max temp of 125°C. Why do you need to tell it your ambient temperature? Something smells fishy here. Has any of the many toyers ever measured the real temp at the tip and compared it to the display? How does it react to heat flow out of the tip?
>the actual modification required to turn this into a 1kW iron would actually be fairly minimal
Sure. Heater resistance is about 9Ω and 120^2/9 = 1.6 kW (at 13.3 A). For a second or three.

Welcome to Absurdistan.
>>
>>1230783
>>1230735
Actually those tips(and T12 ones) have thermocouple in series with the heating element. Also 150v will probably cause the tip to blow up, if you want more power you can go for JBC 250W tips.
>>
hey, it's my first time posting on this board.

I just wanted to know what would be the best way to step up 5V DC to 12V DC using a USB port as a power source. Obviously I could just hook it up to the power supply, but what's the fun in that, am I right?
>>
>>1230730
Remember that each of those ADCs has 8 channels. The board has pinouts for everything, look at how STM32CubeMX works, you can assign most pin roles in code.
>>
>>1230732
If you're running it at 12V then no wonder you can't use it normally, as it takes ages to heat up @ 12V. Just use 20-24V and it will heat up almost instantly allowing you to lower the standby delay. I soldered with it for a couple of hours, it's easy if you don't solder for an hour continuously and you obviously don't.
>>
>>1230863
Boost converter, something like http://s.aliexpress.com/neYveiiU
>>
>>1230869
Oh thanks, I thought about that as well, I just thought there might be a better way (not that it's bad but there are always alternative). Where could I learn more about that, so I know which components I should get, since I want to assemble one myself
>>
>>1230872
Determine how much voltage and current you will need, find a controller chip with suitable parameters, find the datasheet for it, and follow the reference schematic from the datasheet.
>>
>>1230794
>thermocouple in series with the heating element
Interesting, didn't know that. It explains the 0.1% tolerance of R1 and R2 in the TS100 circuit diagram.
>T12
Is this the kind where you throw away the entire heater just to replace the tip? Seems wasteful at first but if they live long enough it may be acceptable.
>>
>>1230876
what do I need the controller chip for? sorry, electronics noob here
>>
>>1230877
That gives you way better heating speed tho. And chinese fake tips don't even cost that much, around 2-4$/tip. TS100 tips are more expensive at around 9$

>>1230880
Google boost converter to learn what the controller does. A common example would be MC34063.
>>
>>1230885
alright, thanks mate
>>
>>1230866
>Just use 20-24V
Car battery, solar powered environment. Maybe my laptop adapter? I actually do solder for an hour continuously, and more. When you populate dozens of boards the thing never sleeps because you need the heat. I have a 12V 8W iron for SMDs, but it's too weak. The 17W are just right.
>>
Any tips for desoldering components (mostly through-hole), preferably without damaging the board or ruining components. Right now I am using solder braid and it never seems to work that great. I start by apply fresh solder to the pad or lead, then I place the braid near the pin/lead of the component. I place my iron on top of the braid and apply some light pressure to sandwich it up against as much of the pin and pad as possible. Usually it seems to take a long time for solder to go into the wick and once it does and I pull the iron away I find that only the solder on the surface was taken up so I have to go through multiple applications to get it all. I typically leave the iron on the component until the flux from the braid begins to burn off and emit smoke. I'm worried that in the process I apply too much heat to the board and components.

My iron is a Hakko FX-888D typically set to about 750F or ~400C. I have a chisel tip that is slightly oxidized but not in awful shape. I usually use 63-37 tin-lead solder in stuff I make but when replacing components in test gear or other electronics I'm most likely dealing with RoHS compliant solder unfortunately. I have no flux pens at the moment but my solder and solder braid does contain flux.

Any tips or advice is appreciated. Am I doing something wrong or does solder braid just suck? Should I invest in a vacuum desoldering pump or at least one of those cheap solder suckers?
>>
>>1230940
You can get a pack of desoldering needles for less than 2$, they're quite helpful, especially for DIP ICs.
>>
>>1230880
Well you know a typical boost converter works by rapidly switching a component, typically a MOSFET, on and off. During the on time the FET is shorting an inductor to ground causing a large magnetic field to build up. When the FET is off the magnetic field collapses and the current is passed through a diode to charge a capacitor, the cap cannot discharge back through the diode so the voltage on the cap builds up.
The MOSFET is typically controlled with pulse-width modulation, now you could just take signal generator apply a square wave with of a particular frequency and duty cycle to get a rough output voltage you want and call it a day but it won't be stable. When you apply a load that voltage will change. This is why we need the controller chip. By adding the chip and a simple voltage divider feedback network this allows us to get a stable voltage regardless of the load (well so long as you aren't trying to draw more power than your boost converter can deliver).

The controller chips are not hard to use. I made a boost converter with the MC34064 that can output anywhere from 70V-270V from 12V in with a maximum output current at 170V somewhere around 40mA so about 6.8W of power.
>>
>>1230947
Great explanation, thanks man
>>
>>1230940
Solder braid is a pain in the ass. The best way I've found is to use a spring-loaded solder sucker. You can also get purpose-built desoldering irons that have a bulb that sucks air through a hollow tip. Those are quite useful.
If the components are resilient to heat and you don't care about the board, you can use a small blowtorch flame to heat a lot of pins at once. A hot-air station can also be good for that with less of a chance of damaging things.
>>
>>1230750
The STM32F4 does seem to have the necessary USB and I2S hardware. It just seems like a bit of overkill for stereo out. If you wre doing any more than stereo, such as a pro audio interface with like 12 ins and 12 outs, the STM32F4 would be interesting.

>>1230783
>Why do you need to tell it your ambient temperature?
Cold-junction compensation. Thermocouples can only measure the difference between their two leads. You need to compensate that.

>>1230940
>I have no flux pens at the moment
>Am I doing something wrong
Yes, invest in a solder sucker and some liquid flux. There's almost no such thing as too much flux, almost.

>>1230947
>output anywhere from 70V-270V from 12V
Don't tell bikeanon
>>
>>1230941
>>1230964
>>1230978
Alright thanks for the replies, I'll probably go invest in some flux pens, desoldering needles, and one of those spring loaded solder suckers. Fortunately it's all pretty cheap. If I'm still having trouble then I'll start considering hot air guns and vacuum desoldering pumps and the like.
>>
>>1230783
>a max temp of 125°C
Literally unable to work on lead-free solder, which I'm sure many have done. Being able to tell the ambient temperature is good because it allows the tip to be calibrated. The point of turning it into a kW iron is that the thermal resistance between the heating element and the thermocouple are small enough such that the controller can still limit the power to the tip before it overheats, the only question is whether the extra voltage would be too much for the insulation, and whether the heat would really travel fast enough to the thermocouple.

>>1230978
Don't worry, I needed a sine wave.
>>
>>1230980
Pretty sure flux pens are shit, I use a tub full of rosin-solder and it's great, if a little tough to apply small amounts from.
>>
File: TS100-thermo.png (18KB, 460x709px) Image search: [iqdb] [SauceNao] [Google]
TS100-thermo.png
18KB, 460x709px
>>1230981
>ambient temperature is good because it allows the tip to be calibrated
How so? The built-in thermometer circuit is much more accurate (2%) than the user-supplied value will ever be. The manual says the device should be at ambient temperature when you click in your numbers.

The image shows the relevant excerpt from the circuit diagram:

B is system ground, A is the switched DC input that feeds the heater, TS is the tip/heater, U6 (TMP36) is the thermometer circuit and U1 (SGM8551) seems to have a very precise gain setting. R31 and R32 lead to two ADC inputs of the controller (a third one measures the actual DC supply voltage). The tip/heater has exactly two terminals. The outer metal part (the one that burns your fingers) is only connected to the ESD grounding screw.

Where's the thermocouple and what temperature difference does it sense? It can only be part of the tip/heater ensemble.
>>
>>1231015
Thrermocouple is in series with the heating element and the ambient temperature sensor is for cold-junction compensation.
>>
File: TS100-supply.png (37KB, 1146x398px) Image search: [iqdb] [SauceNao] [Google]
TS100-supply.png
37KB, 1146x398px
>>1231023
This has been said before but there is no ambient temperature sensor. Ambient temperature is unavailable to the device as soon as you take it in your hand and even more so when you switch it on. It _was_ available before you picked it up and it can change anytime without prior notice.

Whatever, here's the interesting power circuitry. The two anti-serial PMOS FETs (CJQ4953) have a max Vds of 30V and the only way to switch them ON (as far as I can see) is by switching Q2 on which in turn is only possible via C10, a 10nF capacitor that goes to a digital port of the controller. How the f does that work.
>>
>>1231060
Actual air ambient temperature isn't relevant, just needs to be close enough to the temperature at the cold junction.
>>
>>1231012
I have a flux pen at work. Buy a few, you can refill them with a syringe and needle but eventually the felt tip wears out/mushrooms to the point where you can't clip any more of teh whiskers off and still have it be effective.

Our manufacturing/tech team uses 3oz squeeze-bottles full of rosin with a needle shoved through the cap. Enough rosin to last a while, and still able to pinpoint the parts that need rework.

But they're doing orders of magnitude more Pb-Free soldering than I am.
And I generally use leaded solder for my work. Much easier to work with and none of my stuff has to fall under RoHS.
>>
>>1231060
Well that's not quite correct, by looking at D4, R11, and R29, shouldn't the FETs be on constantly? When Q2 is off, the voltage at the gate of the FETs will approximately be Vin, and when Q2 is on, the voltage at the gate of the FETs will approximately be Vin/2. I think the FETs will be on regardless since Vin is at least 12V, and as you said there doesn't appear to be any way to toggle Q2, or at least not without a forced 50% duty cycle. Assuming U2 and U5 are just part of the switching logic-level power supply to get 3.3V to the µC and OLED and 3.9V to the temperature sensor U6 and opamp U1, I simply can't see how the switching part works.

>>1231069
Shit, one of those needle-tipped modelling glue bottles would be perfect with runny enough flux. Maybe I can shove my jelly-flux in there with a small heating element.
>>
>>1230603
Is he planning on soldering the inside of a HDD? What a madman.
>>
>>1231060
>Ambient temperature is unavailable to the device as soon as you take it in your hand and even more so when you switch it on
Correct. The temperature of interest is that of the cold junction of the thermocouple, which is approximated in absolute terms by the TMP36.

>>1231069
I use SAC305 regularly like the prissy little green bitch I am. Also, I figure it's good practice in case I have to sell my skills come the next scheduled global financial crisis. It beats oral.

>>1231078
>the voltage at the gate of the FETs will approximately be Vin,
They're high-side switches. As far as they're concerned, Vin = 0V.
>and as you said there doesn't appear to be any way to toggle Q2
Sure, C10 is there to pass edges, not levels. C10 enforces that Q2 doesn't stay on continuously in case of a microcontroller failure, thus not setting the building on fire.
>Shit, one of those needle-tipped modelling glue bottles would be perfect with runny enough flux. Maybe I can shove my jelly-flux in there with a small heating element.
Real pros thin it with IPA to the point of dispensability.
>>
>>1231088
>to pass edges, not levels
Pretty sure that still means the net duty cycle through the cap can't be anything but 50%, which doesn't seem too useful for running like a speed controller. In the very least the duty cycle couldn't be driven over 50% for very long, since I guess Q2 will probably remain "off" when the voltage across C10 is at a steady state. All in all it looks like a particularly less-than-ideal way of going about it, especially when you could probably add some fairly simple protection circuitry in case the FETs blow closed or the µC "malfunctions". All it would take would be a transistor or a relay driven off a flip-flop from a comparator on the thermocouple that goes open-circuit until an internal reset button is pressed if it exceeds more than 50° above max. Or something much simpler than that, I don't see it as being much of a threat anyways.

But still, what is the second, reverse polarity FET doing?
>>
>>1231118
>net duty cycle through the cap can't be anything but 50%
You'll notice there are bleeder resistors pulling the Q2 base to ground. Think pulse frequency modulation.
>since I guess Q2 will probably remain "off" when the voltage across C10 is at a steady state
Correct. However, Q2 discharges C24 when a positive edge is received, which charges back up through R29/R11/D4. Even with a 50% maximum pulse with, which isn't the case, above a certain pulse rate, Q1A/Q1B will remain ON 100% of the time.
All in all, it's a pretty good failsafe against programmer error, but the FETs themselves could still blow short.
>a comparator on the thermocouple
Easily said, but getting that voltage is difficult in this design.
>But still, what is the second, reverse polarity FET doing?
I'd guess they were using battery charging MOSFETs and just borrowed the connection topology in its entirety. Maybe it was a stab at making it run on low voltage ac? If that STM32 ADC accepts a -4V input, that may indeed be the case.
>>
which direction do i i put an electrolytic cap if i want to filter noise on a -12v rail

it seems intuitive that the anode would be connected to ground in this case? please no bully
>>
File: DSC00009-L.jpg (213KB, 800x600px) Image search: [iqdb] [SauceNao] [Google]
DSC00009-L.jpg
213KB, 800x600px
>>1231145
- lead is always the lower voltage; + is always the higher one.
Unless you want >pic related.
>Customer stated they "heard it pop and stuff started leaking out."

+ to ground, - to -12.
>>
>>1231139
>Think pulse frequency modulation.
Changing duty cycle by changing frequency? I never considered it, but that's actually somewhat clever.
>the FETs themselves could still blow short.
In which case the safety mechanism does nothing; there'd be a direct short from Vin to Ground. But in terms of people being able to program the thing themselves, it's a pretty good idea.

You're right that both FETs are in a single SOP8 package, though they aren't permanently linked as they are in the circuit, so there must be some reason that they use both here, not to mention why they use that package in the first place. I'm thinking that when the system is solely powered by the 5V from the Micro USB input, the 3.9V switching regulator is powered through D1, and it also puts a high signal behind the two FETs, meaning the sense current will not only still flow though the element and series thermocouple, but also from 3V9 through R26, the FETs, R20, and R21 to ground when the FET is enabled. The voltage between R20 and R21 is also fed into what I assume is an analogue input for the microcontroller, so I guess it measures the voltage at Vin for comparison with the current through the element? In proper operation the voltage Vin would be divided by 8.5 and measured by the controller for calibration purposes, which I guess explains why the voltage into the element itself isn't regulated.

>>1231149
wew
>>
>>1231149
thanks much!
>>
>>1231157
>why they use that package in the first place
Form factor? It could be a TSSOP-8
>it also puts a high signal behind the two FETs
P-channel devices need a negative gate-source voltage to pass current, so that doesn't help. R26 might just be a pull-up resistor for presence/continuity sensing so that a broken element reads way high when the pass transistors are off for sensing, instead of way low which could happen under a cold start or after a change of element.
>>
>>1230864
Oh yeah, you just reminded me of ADC channels, so I could just scan all the input channels in series, and that way I only need the one ADC pin. Which communication protocol would be best for that by the way? It would also be nice to have some knobs which control volume, but 8 channels would be enough for stereo at least.

>>1230978
Just the stereo output sonuds find, but I would like to try and play around with more than just stereo inputs. Compared to the teensy, I think that discovery board just looks a little bit nicer and harder to lose it or something. Not to mention I kind of want to do more ARM programming rather than AVR programming.
>>
>>1231173
I think R26 is actually for temperature measurement. Since the element always has 3.9V across it and R26, the change in resistance of the element as its temperature changes will change the voltage at A, which will be measurable as the temperature. At least this works when Vin is not connected, hence the whole "measuring rest temperature for calibration" thing.

The matter of the FETs means that the circuit is 99% certainly designed to put current both ways through the FET array, because otherwise the extra diode drop would just decrease efficiency, and they'd have left it disconnected or at least all tied to ground. By high signal behind the two FETs I meant behind R26, behind their sources and drains respectively. The only way I can see current going from left to right is from R26 and to either D4 or R20 when the thing is running off USB power alone. Since R20 just leads to a voltage measuring analogue input, which shouldn't change since it's being regulated at 3.9V, the purpose must be for biasing the R11-R29 divider for the purpose of triggering the FETs in the first place, which to me sounds like circular logic. D4 also leads to U2's EN and Vin, but those will already be being powered by the 5V input. I don't get it.
>>
>>1231078
>Shit, one of those needle-tipped modelling glue bottles would be perfect with runny enough flux
I use pure rosin dissolved in ethanol in those bottles for really cheap flux.
>>
>>1231193
>I think R26 is actually for temperature measurement
With all the high precision components in Pic related in >>1231015 also supplied by that 3.9V supply, it doesn't wash that they would use a single resistor, and not even a precision one, for temperature sensing. So, is there a thermocouple in the tip, or is the element just strictly heater wire?
>99% certainly designed to put current both ways through the FET array
FETs pass current both ways, just slightly less efficiently in the other.
>decrease efficiency
Not sure they cared if they expected these to run off large batteries or wall adapters.
That's why I'm thinking that the circuit *may* also run off ac, if equipped with a micro that better handles highly negative voltages on its ADC inputs.
>The only way I can see current going from left to right is from R26 and to either D4 or R20 when the thing is running off USB power alone
Which also requires that the MCU be pulsing Q2. If current passes from left to right, it's going to be moderately choppy, especially through a 39k resistor with all those other loads on the other side.
>I don't get it.
That's what I'm saying, senpai. The heater is a low-impedance device, as is a thermocouple. 39k would be swamped by them. Idle calibration would merely involve not generating a heat gradient.
>>1231187
That's fair. AVR programming has its time and place: namely, cost-consciousness. ST's chips cost like $11 each in singles, which is not very experimenter-friendly compared to low-end tablet SoCs.
>Which communication protocol would be best for that by the way?
An 8-channel audio-quality DAC is not the most common bird. I suggest you find your ADC/DAC first, then work around that. Chances are the ADC/DAC will speak I2S or AC'97, or something close enough to it. Parametric selection guides are a great way to piss away hours.
>>
>>1231207
I meant an 8 channel ADC would definitely be good enough for stereo input, sorry. I would be happy with just a 2 channel DAC.

I have been wondering on how interfacing with an external DAC or ADC works. Do I just bypass the microcontroller and just send the data through the USB for an input and just to a jack for the output?

Secondly, I've been wondering about the speeds needed to actually sample and send the audio at. Are there any drawbacks asides from power consumption than just increasing the clock rate with a crystal?
>>
>>1231207
AC might work going by the double FETs alone, but since D4 will only conduct on one half-cycle and the FETs will only trigger with one polarity in gate voltage it doesn't make sense. I guess it's just a little oddly designed in that respect.
>>
File: T12_tips_handle.jpg (81KB, 1480x800px) Image search: [iqdb] [SauceNao] [Google]
T12_tips_handle.jpg
81KB, 1480x800px
>>1231207
>is there a thermocouple in the tip
Yes, directly in contact with the tip. That's the principle behind these T12-type soldering irons. They are extremely light-weight and have almost no 'thermal mass' to draw from like a conventional tip has, so fast regulation is mandatory. When you look up 't12 soldering iron' you will be flooded by Alibay et al. with dozens of models, kits and regulating+displaying power supplies.
>>
File: motor cutaway.jpg (24KB, 494x455px) Image search: [iqdb] [SauceNao] [Google]
motor cutaway.jpg
24KB, 494x455px
I've got a brushed electric motor in a cheapo shaver that I want to spin faster and/or with more torque(?)
The original cable is missing but the shaver says it needs 3V which seems a bit low.
I've tried 3, 4.5 and 6 volts with an amperage of 300mA. At higher voltages it spins faster but it stops at 7.5V.
Still at 6V the blades don't spin fast enough to shave effectively.

Should I increase the amperage?
Will it make the motor spin faster or with more torque?
Or is it useless? I don't think I have an AC/DC adapter that goes higher than 300mA so trying this idea out is going to be a pain.
>>
>>1231396
If you have or can get a meter, measure the voltage the actual voltage applied while the shaver is running.
The motor may be pulling the voltage down due to low available current (mA)
>>
>>1231396
>the shaver says it needs 3V which seems a bit low.
two AA batteries in series would supply 3vdc at more than 300mA
give that a try
>>
File: tenor.gif (2MB, 294x220px) Image search: [iqdb] [SauceNao] [Google]
tenor.gif
2MB, 294x220px
>>1231396
>amperage
>>
https://www.youtube.com/watch?v=uD6hoAzjQqQ

>7:40
>we can copy tesla's inverter at home

does anyone have a tesla?
>>
>>1231468

8:22
>>
>>1231460
English isn't my first language
Don't assume spelling errors are the result of a lack of intelligence
>>
Does anyone know how to calculate reference frequency in an op amp?
>>
>>1231208
>I have been wondering on how interfacing with an external DAC or ADC works.
I2C, SPI, I2S, the usual. You really should go stare at some of those parametric selectors on, say, TI's site to see just some of the options you have. In general, you handle USB requests and repeatedly: initiate conversions (for input), read input buffer or ADC, write to DAC or output buffer. Better to have 2ms of ring buffer for each channel, as you'll be read/written every 1ms and not necessarily at the exact same cycle every time.
>Are there any drawbacks asides from power consumption than just increasing the clock rate with a crystal?
Yes, the USB subsystem needs a highly accurate clock. You might also lose accuracy at the ADC.

>>1231215
>FETs will only trigger with one polarity in gate voltage
Oh, good catch. Yep, just weird and/or lazy, or maybe that part just didn't work on the bench that way.

>>1231477
It's a running joke around here. The preferred term is "current".
Anyway, if you want to change the parameters of the motor, you'll most likely need to rewind it or replace it. Also, be careful not to "surprise" any batteries that might be present.
>>
>>1231505
read the datasheet
>>
>>1231477
amperage is totally correct
the rule is:
if voltage then amperage
if tension then current
>>
>>1231534
what about ohmage?
>>
>>1230940
no-brand solder wick straight up doesn't work. it doesn't wet out when you put the iron on it, and even if you flux the shit out of the joint it doesn't wick up more than a couple millimeters.

chemtronics or goot pretty much instantly wet the joint out, and the solder will soak about a centimeter along the wick in both directions. stick to 2 or 3mm rosin fluxed wick for your hakko, it's plenty absorbent enough to empty a through-hole in one application. fatter braids of wick would be too heat-sinky for your thermal mass; you'd have a bad time.
>>
>>1231532
i dont have a data sheet. only a circuit diagram with resistors and capacitors
>>
>>1231534
>tension
That's not a common one.

My rule is:
if measured in multiple amps (1A<I<1000A) then amperage is fine, like on a welder. The same also applies to voltage and wattage within a closed context, such as household appliances where voltage and wattage are fine, a welder where amperage and voltage are fine, and large radio where wattage is fine. Resistance is almost always in a context where it can vary either between very high and very low, meaning its more important to state the unit of resistance, meaning calling it ohmage is pointless. Voltage is pretty universal because pretty much everything you'll encounter has a potential between 1V and 1000V, but if dealing with MV I might be tempted to state "electric potential" instead.

As far as I can understand, "voltage" is the magnitude of volts, and hence can be represented with a number without a unit, since the unit is already within the value's descriptor, and the same can apply to any unit as long as it doesn't cross into the domain of another unit, like tonnage, kilogramage, meterage, and sensual-massage.
>>
>>1231538
What's it called? If it's just a "generic opamp" then you can assume ideal characteristics like infinite gain, infinite input impedance, etc. but that isn't too helpful. If there's a part number, then there's a datasheet online. Also pics help.
>>
File: 20170821_210018.jpg (4MB, 4032x3024px) Image search: [iqdb] [SauceNao] [Google]
20170821_210018.jpg
4MB, 4032x3024px
>>1231541
Integrator circuit
>>
>>1231508
Sorry for asking so many questions, I am a little bit new to actually doing projects myself. I kind of want to make an order, but I need to spend a bit more than just the board to get the free shipping, so I just want to try and get all the components that I might need in the future.

I didn't realise that they were inaccurate. You think the 16MHz internal clock of the chip would work fine?
>>
>>1231547
So what do you mean by reference frequency? It's not like that thing is an oscillator.
>>
File: multimeter!.jpg (2MB, 2448x3264px) Image search: [iqdb] [SauceNao] [Google]
multimeter!.jpg
2MB, 2448x3264px
>>1230603
FYI if you need a decent meter for cheap this one is excellent:
https://www.banggood.com/ANENG-AN8002-Digital-Ture-RMS-Multimeter-ACDC-Current-Voltage-Frequency-Resistance-Temp-Tester-p-1145700.html?p=2T111812202419201707
I tested it against the fluke at work and it's spot on, measures very small capacitance pretty good too.
>>
Hey /ohm/. I have some newbie questions about lithium ion cells. I have 4 unprotected Samsung 25R 18650s that I want to wire in parallel to feed a buck converter. The converter takes between 2.9 to 5 volts in and outputs 5 volts. The max amperage it can draw is 5 amps. I'd like to attach a protection and charging circuit to the 18650s without blowing myself up, but the guides I find online are limited to pairing a single protection circuit with a single cell. The guides for lithium ions in parallel are also somewhat sparse and filled with contradicting opinions on how to hook parallel cells together safely. I was wondering if anyone here had experience with this kind of thing that they wouldn't mind sharing.

From what I've read online, one can connect lithium ions in parallel if they are from the same batch when made. While I can't say if they were from the exact same batch, I ordered them all at once, so hopefully the cells are close and at the very least have experienced the same wear and tear. This should mean that the cells will be relatively balanced.

I guess I'm asking how to hook up the protection circuit. Should I put one protection/charging circuit in series with each battery and then connect them in parallel? Or should I connect the batteries in parallel and only use one protection circuit between them and the converter? If this is the case, am I just putting my faith that none of them will short seeing as nothing would be protecting them from each other?

Also, if anyone has recommendations for redesigns or particular protection ICs, I'm all ears.
>>
>>1231561
What are the leads it comes with like?
>>
How do I power a 5v Pi and a 12v display off of a 20v battery?
>>
>>1231624
buck converter for each or voltage regulator for each? I think voltage regulators burn off the excess power so a converter might be more efficient and heat up less. Pololu sells converters. I just bought a step up one myself.

>https://www.pololu.com/category/131/step-down-voltage-regulators
>>
>>1231631
Hot damn, thanks a bunch, that's exactly what I needed, cheap, and in plenty of options. I've asked before and kiiinda gotten helpful answers, but this is something-is-going-in-the-cart-tomorrow.
>>
>>1231635
No problem man. Best of luck
>>
>>1231551
It's all good m8.
>I just want to try and get all the components that I might need in the future.
That's the kind of thing aliexpress is great for. A few vendors have passives assortments of many varieties: through-hole and SMD resistors/capacitors/inductors in various case sizes, voltage regulator kits, LEDs by the bucket...
>You think the 16MHz internal clock of the chip would work fine?
The Discovery board should be ready out of the box. Separate converters are a separate matter which will need clocking too, some requiring an external clock (a plain old crystal will be just fine as long as it's the right value) or the I2S clock, which the SoC generally provides configured with one or two registers. Many SoCs will even DMA the I2S port to/from your choice of RAM location.

>>1231568
Surely you meant a boost converter?
>attach a protection and charging circuit to the 18650s without blowing myself up
Of course, use battery cases or a spot welder. Don't solder or bad things can happen.
>one can connect lithium ions in parallel if they are from the same batch when made
As long as they've had about equal wear and tear, they're fine.
>Or should I connect the batteries in parallel and only use one protection circuit between them and the converter?
This is fine. Your pack will be limited by the weakest cell, so as long as your cells are matched, and it wouldn't hurt to data-log a charge/discharge cycle on each of them to check, you'll be in great shape. Be sure your protection board is rated for that much current draw. https://www.aliexpress.com/item/1S-7A-4-2V-li-ion-BMS-PCM-battery-protection-board-bms-pcm-for-lithium-LicoO2/32428487954.html
If you want to design down to the board level, the DW01 protection IC is common. Just add MOSFETs and a couple of passives. The cheapest charging IC is the Top Power TP4056, which is good enough for a single cell. For larger packs you'll want something heavier like the TP5000, if you can read Chinese...
>>
>>1231635
My recommendation: get a bunch. They are crazy handy to have around.
>>
>>1231646
For sure, I'll pick up at least a spare of the couple I use, just enough that next project doesn't need to wait on one more order.
>>
>>1231645
Damn man. You've saved me a bunch on time, money, and limbs. Thanks!
>>
File: tensión.jpg (33KB, 430x530px) Image search: [iqdb] [SauceNao] [Google]
tensión.jpg
33KB, 430x530px
>>1231539
>That's not a common one.
I know, and I adapt to that and call it a voltage because when in 'Rome'.. It is still linguistically inconsistent and outside of the anglosphere tension (or something to that effect) is in no way uncommon. The common denominator is the mechanical analogue (pressure, current, flow) and voltage is clearly the odd one.

Combining the name of a unit with the suffix -age (often denoting a collective quality) is in itself a rather helpless construct. Voltage and amperage follow the same linguistic pattern and accepting one while rejecting the other appears arbitrary. But hey, it's langu-age after all..
>>
>>1231645
Looking at data sheets, I can sample at 128*fs, and I would like to end up sampling at 192kHz, because if I'm going to all this effort, I might as well make it nice, so I can use a 49.125MHz crystal for the converters. I am a little worried that I'll find out something that I won't be able to sample all channels at that rate or something. I think I've settled on 24 bit external 8 channel adc and an external 2 channel dac, but I'll try and build it using the internal DAC and ADCs at first.

I've read a little bit that internal microcontroller oscillators aren't very accurate and that an external crystal for the microcontroller should be used for USB communication, so I'm just not so sure. Would the development board and the microcontroller be capable of sending all that information through USB?

Yeah, I'll get all non IC or important stuff online like ebay or aliexpress, like jacks and the sorts. I also see crystals on there for pretty cheap, are those trustworthy at all?
>>
>>1231781
Bah, I was looking at the wrong part for the ADC, I'll have to do more research then.
>>
>>1231534
>the rule is:
Who's rule is that?
>>1231535
What about secondage for time? Stupid isn't it.
>>1231539
Nothing wrong with tension other than its not specific. Same reason partly nobody says potential difference outside the classroom, besides the fact it's cumbersome.
Nobody is going to think you are feeding an elephant buns if you say current
> my own rule
Oh so everyone should just make up their own rules should they about when to use amperage and when to use current? Well here is my rule, don't use amperage at all.

You still need to put the unit of you write voltage of course you do what a dumb thing to say.

>>1231780
>linguistically inconsistent a
Opening s fucking can of worms there my friend.

If anyone wants a real set of rules, not ones made up by people on here, check the SI units, voltage-volts, current-amps.

As they seem to have never heard of SI units and don't play well with others I can only assume amperage is one of those Americanisms invented by some fat lazy stupid idiot.
>>
>>1231780
The problem with "potential", "tension", or "pressure" is that they all have to be specified as referring to the electric vector field, but in some way this also applies to current and resistance. I think the most scientifically accurate term to use would be "electric potential", with "electric pressure" next since potential is proportional to pressure in the flow analogue. But even "electric potential" isn't perfect, because it's referring to the potential energy, and unless specified that this energy is per electric charge, it could easily be assumed that this potential refers to the amount of energy stored in an electric field.

Looking deeper, since the analogue for "q" in an electric field is "m" in a gravitational field, we can say that since voltage V = E/q, that some "gravitational potential energy per mass" is equal to E/m, but this isn't terribly useful as we don't have materials with differing "gravitational permeability", nor do we have materials that conduct gravitational charges, nor do we have (at least not at the moment) gravitational charges of opposite polarities. The idea of applying flow properties to gravity doesn't seem to work with our current understanding of gravity, at least from a classical point of view. In any case, its units would be "(m/s)^2". Magnetic fields don't have useable magnetic charges, so it stands to reason that there's no analogue for voltage in magnetic systems. They're also not technically force fields in the first place, though I think that may be based on the assumption that there aren't any magnetic monopoles. So since there aren't any other examples of force fields apart from gravity fields and electric fields, we don't have an analogue for voltage to base any better terminology off. Just trying to think of the gravity analogue for a capacitor makes my head spin, so does wondering what a negative mass would do in "E=mc^2".
>>
>>1231785
Amperage is used by welders everywhere, call it an americanism if you want, but it works just fine. I'm an SI guy and am fine with calling it voltage, but I'm a scientist more than an SI sheep and I'd like to get to the bottom of this vector field thing.

>>1231786
Now let us not rule out the idea of comparing electric fields to gravity just yet, as we still have the actual field strength to think about. A gravitational field strength, being equal to the force divided by mass, is the local acceleration, and equally force divided by electric charge is equal to electric field strength. Voltage is electric field strength multiple by distance, so we can also think of gravitational potential (as in potential energy per mass) as its acceleration multiplied by distance. If we think about this in terms of masses, two positive masses of similar magnitude near each other will have a near-zero acceleration/force in between them, and hence a field strength of near-zero. If we allow for negative mass, we can imagine two oppositely charged masses of similar magnitude, and the closer they get the more their local accelerations add up, meaning the field strength between them is greater, which is reflected in (acceleration) = (gravoltage) / d. But that was thinking about the field strength in terms of the masses and charges, not volts/gravolts, the proportionality between which is the factor of energy. But since energy doesn't remain constant as the charges/masses are moved closer or further away, this certainly is no longer an intuitive explanation.

Now if electric potential was accurate then I'd go with that, but since potential would refer to electric energy, I think electric pressure is the best fit. And electric tension would imply force. So I guess if I was a real arse I'd describe the energy stored in a capacitor as electric potential, the force felt by the plates as electric tension, and the voltage across them as electric pressure. Fun!
>>
>>1231612
they are probably not 1000v rated, but they are silicone insulation from what it feels like. They have the cat 3 tip shields that come off.
>>
At least we can all agree that the proper symbol for voltage is U.
>>
File: included leads.png (267KB, 603x604px) Image search: [iqdb] [SauceNao] [Google]
included leads.png
267KB, 603x604px
>>1231612

If you click his link you can view all that's included for yourself.
>>
>>1231788
>Amperage is used by welders everywhere
No it's not.
It's not used outside of English speaking countries.
You could argue it's used in the UK but I have never heard it nor seen it written on a welder that wasn't a Lincoln.
So I'm sticking with Americanism.
>>
>>1231791
>silicone
That's enough of a selling-point for me, I'm sick of PVC garbage.

>>1231797
Are those standard leads? I've never touched a pair, hence why I asked what the leads were like as opposed to what they looked like.

On this topic, are there any particularly rigid but precise alligator clips I could use as my "ground clamp" when I'm probing voltages with my red lead? Those parrot-clips or whatever the standard oscilloscope leads are look ok, but I'd rather have something that can carry a little more current and not interfere with resistance measuring by much. The only alligator clips that I've seen that are more rigid looking than the standard ones are also twice as wide and don't have serrated jaws, and so don't look very suited to clamping onto a PCB test point or resistor lead.
>>
What's the most compact/simple way of implementing a photoresistor-managed switch? I only need to move about 40mA
>>
>>1231788
In other languages 'tension' denotes stress and has the dimension of pressure.
>>1231785
>opening a can of worms
Just briefly, for cursory inspection. I then gracefully closed it again. Because WormsRUs.
>some fat lazy stupid idiot
I'm currently living in a SI-land where voltage and power are female while current is male and where units have no plural form, but I do not dwell in an 'us vs them' mental concentration camp.
>>
>>1231390
Thorough Disassembly of TS100 Mini-soldering Iron Tip
http://www.minidso.com/archiver/?tid-1110.html
This is what the thermocouple at the very end of the heater looks like.
>>
File: globe.png (42KB, 156x156px) Image search: [iqdb] [SauceNao] [Google]
globe.png
42KB, 156x156px
So I got myself into EMI/RFI hell.
I'm building a scanning acoustic microscope. The scanning is done with a pair of NEMA17 steppers doing 1/16 microstepping. As you can see in pic related I kindof underestimated the impact of EMI/RFI. As soon as I enable the drivers I get noise ranging from 3 to 13MHz peaking at ~8MHz.
I've slowly escallated my countermeasures, extra shielding for the electronics, ferite beads ,optocouplers between driver board and the rest and seperate power supplies. Nothing realy helped so far. At the moment the Steppers are connected to the driver board trough ~1.5m of shielded cable. Would it be a vaild aproach to just put the drivers in smal metal cases right next to the steppers? I'm hoping to eliminate the cables as antennas.
Anything else that might be worth a try?
>>
>>1231854
Any connector/pair that use differential signaling should be tightly twisted like cat6

Are the shielded cables grounded too? The shield needs to be connected otherwise it's just a different antenna. Or try it with and without incase your ground is pushing noise back up into your circuits because it's not draining fast enough
>>
>>1231854
>Would it be a vaild aproach to just put the drivers in smal metal cases right next to the steppers?
Certainly, but whether it is enough is another matter.
Can you manage without microstepping / PWM motor control?
>>
>>1231854
First mute the transmitter, then desensitize the receiver. Locate the component that sends on 8MHz. Suitable probe could be a tiny 8Mhz loop rectenna plus headphone and/or indicator, kind of a search coil. All floating/battery-powered, no further connection to anything.
>>
I'm looking to build an electronic keyboard for a synth. I figure on keys making contact along a length of resistors to supply a control voltage, but I'd like to have some polyphony if possible. I'm definitely not going to mess with the proper rigorous way, but I can't think of anything simple to even get a second voice. Any suggestions?
>>
>>1231898
There are pressure sensitive resistors, but they're kinda crappy.
You can also use LED/LDR combinations, where the keys act as adjustable shutters.
If it's a digital synth, you need a processor for measuring the resistances (bunch of multiplexers and internal ADC, maybe) and sending the key data to the actual synth.
>>
>>1231918
>LED/LDR
That sounds like a bitch to calibrate. Working with just one optocoupler for an LFO was a bit annoying. The actual synth is going to be an arduino taking various analog/digital control signals. I'd consider an IC and get the input over serial, but I've never tried it and I imagine that would still call for a proper matrix, none of which do I feel like messing with.
>>
>>1231920
What's your idea?
What do you feel like messing with?
>>
>>1231927
I'm not sure, really. I intended to use an array of pots and switches to relay effect settings to the arduino, which will use mozzy to produce the output waveform. All of that is pretty straightforward; its just the keyboard that's of concern, both electronically and mechanically. Using a midi keyboard would require a shield for the arduino on top of another library's worth of data parsing. I was hoping for some configuration of the resistor series and switches that would allow multiple buses for multiple voices, but it seems like any attempt would have them interfering with each other unless I build a matrix.

Really I think I'll just settle for monophony and not worry about it. Making keys with hinge and spring is pretty straightforward, right?
>>
>>1231867
>Are the shielded cables grounded too?
jea that actually was the first thing I did.wasn't better before.
>>1231882
>Can you manage without microstepping / PWM motor control?
Not for now if I get desperate enough I'll change my hardware to work with half steps and be static while I do measureing.
>>1231892
How do I get/make such a probe?
>>
File: DSC06611.jpg (3MB, 3734x2784px) Image search: [iqdb] [SauceNao] [Google]
DSC06611.jpg
3MB, 3734x2784px
used my tester on that electric plug.. The other end of the incoming cable lands into some Wago connector. Should I use 2 separates connectors ?
Why this Orange left plug don't provide electricity when the right one does ?

thanks for advises & replies.
>>
>>1231930
What about the concept behind the pipe organ: two keyboards, pedals and the register that distributes the airflow to the various 'voices'? Maybe look into how the old Hammond organ did it. Just for inspiration.
>hinge and spring
and switch, sure. Or you can salvage the entire keyboard from some abandoned instrument. I would first develop my key concept and only then look for the functional modules/components required to realize it. Top down first, the bottom up part is always active in the back of your head anyways.
>>
>>1231953
does the left plug work intermittently?
does the left one work with a plug in the right hand side?
>>
>>1231953

Try disconnecting the blue and brown wires from the good plug and plugging them directly into the bad plug, just make sure it works that way
>>
>>1231953
It might just be the lighting but the neutral terminal on the left hand socket looks like it could be corroded, preventing good contact.
>>
>>1231953
the copper conductor of the wires are proobably not supposed to go up into that part of the socket, rather be clamped by the screw in the middle below.
it looks like they are not clamped in, and even if they were it would be clamping the insulation rather than the conductor which is no use.
>>
File: ploog.jpg (41KB, 502x674px) Image search: [iqdb] [SauceNao] [Google]
ploog.jpg
41KB, 502x674px
>>1231953

looks like spring is bent at this point, and is not making proper contact with wire. on the RHS socket, there are 2 wires, so when the spring expands, it hits both wires, left and right.

in any case, this looks like a completely retarded ways to make an electrical contact. it seems to me, the screws should be the ones forcing the contact, not the springs. in short, that whole wiring thing is WRONG.
>>
>>1231980
>that whole wiring thing is WRONG
It's almost suspiciously wrong, like troll wrong.
Insulation is still on the wire where it should have been removed.
>>
>>1231781
>an external crystal for the microcontroller should be used for USB communication
Correct. However, most SoCs use one (or two) crystals to provide a base clock which is then multiplied and/or divided by integers to get other clocks in the system, which will be as accurate as the crystal.
>Would the development board and the microcontroller be capable of sending all that information through USB?
Let's do some math. 192000 * 24 * 8 = 36864000 bits/s, so USB full-speed wouldn't be enough. If the Discovery board doesn't do USB high speed and large (9kB) isochronous buffers, then no. A micro like this shouldn't have much trouble moving 5-6MB/s around, especially if there are onboard RAM and DMA, though you can't be really sure until you design it a little further.
>I also see crystals on there for pretty cheap, are those trustworthy at all?
They should be fine.
One more note, you'll want to be very careful with your board design on the analog portion of the prorgam, but the datasheets will have some suggestions.

>>1231785
>I can only assume amperage is one of those Americanisms invented by some fat lazy stupid idiot.
Bingo. (another stupid Americanism)

>>1231821
From where to where? If you don't need sharp switching action, a simple resistive divider on the base of a transistor with the load on the collector would work. Otherwise, use a Schmitt trigger.

>>1231920
How's your decoupling? You may need some additional small caps across the motor driver supplies to keep from amplifying supply noise.
>>
I'm trying to do a thing in the cheapest possible way (no microcontrollers):

>wait for signal to exceed threshold voltage
>when it does turn on transistor
>wait some time (x hours)
>turn off transistor

How do?
>>
>>1232015
555 timer
monostable configuration
>>
Does anyone here have experience building a theremin, or know a good tutorial?
>>
>>1232018
>555
>hours
Yeah, about that.

>>1232015
ATTINY4, three resistors and a transistor. ATTINY4 is 34 cents in singles from Digi-Key. You might spend that much just for a capacitor with low enough leakage for long timing.
>>
File: 1472990359164.jpg (263KB, 1275x1650px) Image search: [iqdb] [SauceNao] [Google]
1472990359164.jpg
263KB, 1275x1650px
>>1232015
If you insist on no microcontrollers, try the CD4536 which is 55 cents. You're just not going to get even roughly accurate long times without a digital divider.
>>
>>1232018
I considered 555s but squeezing a long enough interval out of it isn't really viable, I'm looking at T~=14400

>>1232041
>>1232049
No MCU isn't a hard rule, I just figured it would end up more expensive. I'll give ATtiny a go, the lead time on 4s was 2 months but I can get 13s for not too much more. Thnx bro.
>>
>>1232058
Better choice, since the tiny13 has an ADC built in, with which you could also watch for the voltage to drop below a limit and turn off the thing early if needed because of a bad battery charging day or whatever.
>>
>>1232062
Sounds perfect, the input signal is actually coming from a thermistor-transistor arrangement which I was having trouble fine-tuning anyway, a bit of analog capability should simplify things a lot.
>>
File: IMG_4468.jpg (323KB, 640x1920px) Image search: [iqdb] [SauceNao] [Google]
IMG_4468.jpg
323KB, 640x1920px
>>1230603
>>
File: cute.jpg (68KB, 500x375px) Image search: [iqdb] [SauceNao] [Google]
cute.jpg
68KB, 500x375px
>>1230603
Where does /ohm/ order their PCBs from?

I've used SeeedStudio in the past, but I'd prefer something in the US with quicker turnaround.
>>
File: icon_stop.png (2KB, 32x32px) Image search: [iqdb] [SauceNao] [Google]
icon_stop.png
2KB, 32x32px
>>1232015
Crystal + RTC?

>>1232098
>tfw no silkscreen waifu
>tfw I'd cover her with my lead solder
>>
>>1232098
How many?
OSH Park is faster than many overseas outfits and the price is somewhat reasonable.
Interestingly, an outfit called BasicPCB has an omakase "value service" where they squeeze your design onto whatever artwork they happen to be working on at the moment, of whatever random material/finish/soldermask/legend color/copper weight, saving about 25% off of the traditional 2-side 62mil green soldermask job.
>>
Bikeanon here, and I think I've got the circuit to both strobe and dim my rear lights. Anyway, I'm running into a problem where the MOSFETs in my simulation drop a lot of voltage between the load and the +2.5V rail, which is less than ideal. If this is because of the 1.5/-1.5 dropout of the opamp I'm simulating with then that's understandable, but I thought these FETs would turn fully on with 4V to ground. Any clues?
>>
Is there an analog way to buffer a frequency?
As in a pulse of some frequency in the audio range goes in, and then later I can throw a switch and send the same pulse back out.
>>
>>1232132
Oh, and V(n001) is the +2.5V rail, V(n013) is the -2.5V rail, V(n012) is the voltage at the top of the load, R6, and V(n009) is the voltage at the gate of the MOSFET M2.
>>
>>1232136
Not really, best hope for not using a microcontroller is an ADC and a series of shift registers that you sequentially send into a DAC to get out again, but that quickly gets complicated if you want high resolution. You could also try switching a bunch of capacitors to store momentary voltages, and release them sequentially by switching their outputs, but it would require a buffer on either end, not to mention a lot of caps. And all that switching would need to be triggered by something, maybe a differentiating amplifier.
>>
>>1232132
You've put your load on M2's source, not its drain.
(Also U5 is redundant but that's another matter)
>>
>>1232132
(also also, there are colored LEDs in the LTSpice part library, iirc)
>>
>>1232098
PCBWay. Cheap enough, quality's always been fine, can have ~4 days from design to boards in hand if using DHL shipping. There aren't any reasonably priced companies in NZ so if I want shit super quick, knife and veroboard it is.
>>
>>1232141
I was thinking of something like putting it through transformer and feeding the voltage into a varicap to adjust an LC circuit to the right frequency for output. Seems like a lot of math though to even figure out if it would work.
Another idea would be to put it through a spectrum of filters to get frequency components and then proportionately drive resonant oscillators matched to them. But that sounds like it'll need either a shitload of components, or will produce an awful approximation.
>>
File: 1503432570736.jpg (71KB, 502x674px) Image search: [iqdb] [SauceNao] [Google]
1503432570736.jpg
71KB, 502x674px
>>1231953
>>1231980
ameritard so I've never seen this connector before
My guess is there is a square plate under the contact and attached by threads to the screw
The white arrow points to teeth which I suspect are on the edge of the hidden plate
Installation is done by stripping the end of the wire a bit
slip the stripped end between the plate (white outline) and the contact
tightening the screw raises the plate which grabs the stripped wire and holds it against the contact
sort of x-ray image attached
>>
>>1232136
Could use a phase locked loop with the control voltage going through an opamp sample and hold circuit.

Wouldn't be super accurate though.
>>
>>1232157
>>1232160
The usual way to get around drift for "infinite" sample and hold is to feed a DAC with an ADC. "Later" is just too long a time for most analog circuits.
>>
>>1232145
Shit, just realised that myself. If I remove the U5 buffer then the 20µA that goes from the oscillator to the comparator is enough to stop it oscillating; the triangle wave there is pretty finicky.

>>1232151
Their forward voltages are not specified at all, and since I'm using a high or low signal to power them directly from a voltage rail with a dropper resistor, it shouldn't be a problem. If I go for a big enough coupe of LEDs (this is for the rear light) then I'll make a BJT-based constant current regulator for them, but not if I'm going for under 100mA for each string of LEDs. I'll certainly have two in series since that means I'll only be losing about 1V in my regulator, but since it looks like the only red LEDs I can locally get are 20mA, a constant current regulator is probably an unlikely solution, and likely to waste excess power when switching constantly.
>>
>>1232190
>20µA input current
LM358s are more like 250nA max. TL06xs are more like 200pA max. You may want to override that parameter of the part or choose a different op-amp to get a better idea of what would happen with real parts.
>make a BJT-based constant current regulator for them
You might not need to. Consider 20mA an average current over short time. The pulse current maxima for the usual T-5 type LEDs are at least 30mA, sometimes much more. You can probably get away with a smaller dropper resistor than you think.
>>
Does anyone know how to massage a wet cell Pb battery back to life? I've read how to bring back SLAs from the edge, but the only thing I've seen on it is Yuasa saying you need to push up to 10A through them. My issue is it isn't holding a charge.
>>
Is there a "best" solder to use, or is it pretty much all the same shit if you're just buying whatevers available at the local Radioshack/Frys/top result on Amazon/whatever? My current roll is almost out, time to buy more.
>>
>>1232219
Best depends on what you're doing, but Kester and Multicore solders are well-respected.
Choose a flux type according to how clean most of your joints are and how much cleaning you can do post-soldering: no-clean for lightly oxidized wires or stuff you can't wash in water or IPA, RMA for slightly aged leads/lugs/boards where a bit of brushing with IPA is tolerable. You could always get some RA flux and apply it to those really dirty joints as and if needed.
Thinner wire is better for SMD work. Thicker wire is cheaper.
>>
File: LED.jpg (7KB, 225x240px) Image search: [iqdb] [SauceNao] [Google]
LED.jpg
7KB, 225x240px
>>1232208
The opamp I'll likely use is probably the LM324N, which surprised me with its 0.2mV low, but since I'll be using a 4 opamp chip it doesn't really matter whether I use it or not. I already own it, so it's just a matter of mocking-up the circuit on a breadboard to see if its necessary.

I'll actually want to use pic related LEDs, though I can't find a datasheet. Since my maximum brightness will mean a 100% duty cycle, I can't just skimp on the dropper resistor. But since the LEDs will probably be over-rated for this purpose, I doubt I'll ever run them at 100%. The website actually states "Typical: 1.7 V, Max: 2.4V" in regards to the LED's forward voltage, meaning I might be able to put three in series and still get significant brightness out of them, which leads to the question of whether three LEDs at ~1.7V would be brighter than two at 2-2.4V. I'm guessing the latter, but even if I buy them there's no easy way to quantitatively measure it, except maybe measuring the resistance of an LDR a fixed distance away from them. Actually I might try that.
>>
File: alinear techard.jpg (89KB, 1090x775px) Image search: [iqdb] [SauceNao] [Google]
alinear techard.jpg
89KB, 1090x775px
Also am I the only one who sees this?
>>
>>1232236
Are those RGB LEDs? That's a really wide range of forward voltage for single colour from any sensible supplier.
>>
>>1232261
That one isn't RGB, two legs are the anode, two are the cathode. I really just want a volt-lumen or current-lumen graph since "minimum forward voltage" doesn't tell you anything, but since I can't find a matching datasheet, I guess there's nothing to it but to buy them and run some tests. They're 120-140° LEDs, so I don't need to aim them in different directions like my original design, which is quite helpful, but they can't be panel-mounted. Since they're going to be behind a diffuse shield anyway, I'll probably just mount them on a bit of PCB.
>>
File: pcbway.jpg (26KB, 970x151px) Image search: [iqdb] [SauceNao] [Google]
pcbway.jpg
26KB, 970x151px
>>1232153
pcbway looks interesting. thanks
>>
>>1232132
Alright, just a couple of questions more. Firstly, will an LM-317T be good enough to act as the voltage regulator that creates the 2.5V ground for the opamps? I'm pretty sure it will be going by the specs, so I guess this question is more "has anyone done it and how well did it work?". I've never actually used LM-317s, oddly enough, but I understand the basic principles of operation, and it looks like I should be using larger values of the two resistors for more output accuracy, not to mention lower quiescent current. I can't just use a voltage divider because the caps required to keep the ground stable with respect to the voltage rails when running a <1Hz oscillator were ridiculous. Needless to say, nobody sells any 2.5V fixed regulators near me.

My second question, since 2N7000s are by far the cheapest (MOS?)FETs I can buy where I am, are they a good fit for the main switching transistor here? For this task the current will be less than 200mA for sure, switching voltage of possibly below 3V, not high frequency, just as you see. I have the option to save 5c a piece if I buy TO-92s instead of TO-220s, but I think I'll buy a bunch of both LM-317Ts and MOSFETs, and having the higher heat capacity will surely help in future projects, and it's also 5c off each if I buy 10 of them. Still, $9.50 NZD for 10 2N7000s isn't that great by the sounds of it, same with $14.30 for 10 LM317s. This is assuming it's handy to have 9 2N7000s lying about.

Also I think I will keep the buffer U2 in place, as it will let me run a small speaker off it as an electric horn, which leads me to my bonus question, are good horn frequencies (500Hz to 5kHz, I guess?) good for LED PWM? I'll probably run a common emitter amplifier off the buffer to get the extra volume.
>>
>>1232236
You'd need to include the FET in your circuit for an accurate comparison. I don't think you'd get useful luminous flux out of them at three deep, especially with a switch in series eating up those precious exponentials.
A photodiode would probably be more linear.

>>1232306
>LM317s as virtual ground
The LM317 can only drain current. In case the virtual ground jumps up, the 317 won't be able to pull it back down. I wouldn't do it. Also, it wouldn't necessarily stay stable between the rails.
>it looks like I should be using larger values of the two resistors for more output accuracy,
No, you need to use relatively small resistors to keep the regulator's own quiescent consumption from causing inaccuracy.
>I can't just use a voltage divider because the caps required to keep the ground stable with respect to the voltage rails when running a <1Hz oscillator were ridiculous
An LM358 is literally half of an LM324. Grab one. It would make a good driver for your resistor-divided virtual ground. It also gives you one more free opamp with which you could make a second tone to feed into your horn for more attention-getting.
>are 2n7000 a good fit for the main switching transistor here?
Rds(on) seems a bit high. I wouldn't try to dissipate 1W in a TO-92. Maybe if you parallelled two of them, but check into your vendor's small power MOSFETs before you do. I certainly wouldn't get ten at local prices just for a 5% discount.
>are good horn frequencies (500Hz to 5kHz, I guess?) good for LED PWM?
Anything above 120Hz should be just fine for LED PWM. Horns don't usually go far above 1kHz. I wouldn't try to take the LEDs too far above 1kHz, either.
>>
I wanna build some guitar pedals (for myself, not for selling). Been looking for parts, mostly pots, switches and jacks. Anyone know if Tayda Electronics is reliable? I'm ordering from Europe, so a lot of stores that are generally recommended on the internet aren't really an option due to shipping.
Also, any advice on what to stock up on?
>>
>>1232333
Oh of course, I can use an opamp buffer to buffer my voltage divider, never thought of that. Don't worry about power dissipation, they also sell TO-220 2N7000s, so I'll probably get one of each, the 220 to switch the LEDs and the 92 to switch the PWM's voltage divider from high to low. Point taken with the local prices, I'll order a bunch of stuff from China after my next holiday when I can afford to wait a month for shipping. Point taken with the frequencies too. Since I'm running no more than 100mA through the FET, I'd have to drop 10V across it to get 1W of dissipation which will never happen, I'd say a maximum of 100mW instead so I could use the TO-92 one, but the look of a TO-220 standing by itself is that of a serious circuit, and there's practically nothing in it.

Taking into account the voltage drop across the switch, I think 3 LEDs in series is out of the question anyway, so I'll get a couple of the red LEDs and see if I need more. So no need for experimenting, which is a little sad. Considering the rear bike light I disassembled a few months ago simply had 5 5mm LEDs, and that I'm not illuminating anything except the diffuse panel in front of the LEDs, two should be more than enough.

Also I'm adding a stereo socket to plug into the horn's speaker and nobody can stop me.
>>
>>1232336

What about Farnell? It's free shipping in the UK and Ireland at least, so if they do charge you extra for mainland Europe it probably wouldn't be much.
>>
>>1232009
Well, one of the workers I'm doing my university project for just gave me a dev board to play with once I mentioned what I was planning to do. Its a Waveshare EVK XA1 with an atmega128a1u (I think, its hard to read the chip) and it has USB built into it too. I'll see what I can do on it with just the internal converters first.

Yeah, I got the wrong IC, I think I'll just go for 4 inputs instead for a final product. I was thinking about that calculation too, but I just wanted a second opinion.
>>
Appropriate thread to ask about audio?

I have a broken active sub, the amp is busted.

I had the idea of buying another same subwoofer, used. It's basically the same price of a new amp for it.

But what I want to do is splice the two woofers together, so I run two from the same amp.

There's no need for left/right, and I don't play loud at all, so it should work, wouldn't it?
>>
>>1232352
>splice the two woofers together, so I run two from the same amp
If you parallel them you will apply 1/2 the impedance the amp was designed for.
If you series them you will apply twice the impedance the amp was designed for.
You already have one busted amp. Why risk another?
>>
>>1232355
It'd be parallel.

The old one simply died of old age basically.

I don't play loud at all so I don't see how going parallel would do any harm to the amp?
>>
>>1232357
Parallel increases current drawn from the amp, it's effectively a constant voltage source.

If you're not turning the volume all the way up you may still be okay.
>>
File: 2N7000-7002.png (48KB, 506x392px) Image search: [iqdb] [SauceNao] [Google]
2N7000-7002.png
48KB, 506x392px
>>1232338
>they also sell TO-220 2N7000s
The Museum of Hoaxes doesn't sell its exhibits. There's only the 2N7002 and TO-236AB = SOT-23.
>>
File: file.png (6KB, 809x144px) Image search: [iqdb] [SauceNao] [Google]
file.png
6KB, 809x144px
Am I reading this right? This uc has 128 bytes of RAM? Or is it somehow 128x8 bytes?
>>
>>1232383

looks like 128 bytes x 4 but I think you only get 128 for general purpose ram. the stack and other shit gets the other three banks.
>>
>>1232383
128 bytes, or 128 x 8 bits.
Just like there is 2k x 14 bits of flash, instead of 28kB.
>>
File: file.png (65KB, 848x768px) Image search: [iqdb] [SauceNao] [Google]
file.png
65KB, 848x768px
>>1232385
>>1232391
Thank you. I was going to be doing 512 sample FFTs on this thing but it looks like I'll have to use my spare 18 series.

Next question, I'm trying to write assembly code for bit banging a WS2812B chain.

#asm
_asm
MOVLW 0x20
MOVWF FSR
MOVLW 0x18
MOVWF INDF
START MOVLW 0xFF
MOVWF PORTB
NOP
NOP
SUBWF PORTB, 1

DECFSC INDF, 1
GOTO START
_endasm
#endasm

4chan's probably going to fuck up the spacing so check it out here https://pastebin.com/c6jphuPF

MPLAB is telling me that the SUBWF and DECFSC lines have syntax errors, and that INDF is defined more than once. Based on the instruction set section of the datasheet, I can't see what's wrong with any of what I wrote. This is the first assembly code I've ever written, so excuse me if this is a stupid question, but how did I screw it up?
>>
>>1232399
> it looks like I'll have to use my spare 18 series.
> (posts pic of 16 series data sheet)
Is it a 16F or 18F?

> MPLAB is telling me that the SUBWF and DECFSC lines have syntax errors,

SUBWF has 3 operands on an 18F. The third one is whether the address is in the access bank or the current bank. I don't remember whether the asm allows that to be omitted.

DECFSC isn't a valid mnemonic on any PIC I know. Should have been DECFSZ?

> and that INDF is defined more than once.
18F doesn't have INDF, it has INDF0, INDF1, INDF2 (likewise 3x for each of FSR, POSTINC, POSTDEC, PREINC).
>>
>>1232465
Sorry, I'm still writing this for a 16. I'll port it once I get this working on my test board.

You're right about DECFSC vs DECFSZ. That fixed the INDF defined more than once error, but I still have syntax errors on SUBWF and DECFSZ.
>>
>>1232507
The ,1 parameter should be the default value. Have you tried removing it? Or replacing it with ,F?
>>
>>1232514
I hadn't tried. Both options work, thank you.

Can you please explain why the compiler wouldn't accept the ,1 but would take ,F or no argument?
>>
>>1232516
The suggestion was just a guess that maybe your compiler expects specifically F/W instead of using the more common approach of expecting 1/0 and having symbols/macros F=1 and W=0.
>>
>>1232379
You're correct, what I was mistaking for a TO-220 was actually a second TO-92 being sold at a different price with a different image by the same shop. Neat!
>>
File: 34-01-R5C.jpg (26KB, 360x270px) Image search: [iqdb] [SauceNao] [Google]
34-01-R5C.jpg
26KB, 360x270px
>>1232528
Does your red LED look like this? I have the Vf/If diagram for my one. The pic you posted has 3 chips, possibly parallel.
>>
>>1232347
USB-to-serial chips can't make class compliant devices. You need a micro that speaks USB natively. U may just be the chip package code.

>>1232338
Right, but also look at the transfer curves and observe that Rds(on) is up in the ohms.
Super-brightness or ultra-brightness round types would be better for your application. If anything, you want the light sources to be somewhat concentrated for better visibility at distance, rather than maximally diffused. Surplustronics LA0084R, for example.
You have much experimenting to do yet.
>>
>>1232635
I think the U might stand for USB, the data sheet says the microcontroller has a 12mbps usb 2.0 device interface.
>>
>>1232666
Hmm, the manufacturer's product page didn't suggest as much, just an FT232 USB-to-serial adapter. If the uC does in fact have native USB, then adding a few resistors and a USB connector should be all that's needed to get it working..
>>
>>1232684
Well, I'm not so sure then, it definitely does have an FT232, but it looks fairly used, so maybe it got swapped out with one which does?

Regardless, it looks like all the documentation and schematics result in dead links for the development board and its discontinued and it doesn't have much support. I might think about scrapping the idea of using this development board.
>>
File: piranha LED white.png (1MB, 1354x1332px) Image search: [iqdb] [SauceNao] [Google]
piranha LED white.png
1MB, 1354x1332px
>>1232560
It does not have three chips, I own a white one from the same seller and it looks as follows, only one bond wire from each side and the pairs of leads are directly connected.

>>1232635
This is a rear bike light, not a torch, I want it to be visible from a wide angle. This Piranha LED emits 20 times more lumens than the 3mm LED, as somewhat inaccurately calculated by assuming the light is emitted in a sharp cone at the viewing angle, instead of calculating it as a normal or sinusoidal distribution with the angle at 50% brightness. 1.5cd doesn't sound like much until you factor in the massive viewing angle. I'll be mounting it behind a diffuse shield, maybe of sanded laminate or acetate, and a 15° LED would just make a small dot on that. I want a rear light that isn't made of blinding little pinpricks of light, which is why I'm choosing to put it behind a diffuse shield in the first place. By maximising the illuminated area on the diffuse shield as opposed to maximising the intensity on a driver's retina, I hope to make the rear light visible from enough of a distance without swamping out the illuminance of street lights on my bike itself. Not to mention they'll be able to see me from more than 15° behind me.

But Jesus, do these Surplustronics LEDs outperform Jaycar's LEDs in both output and price.
>>
What do you nibbas know about earth batteries. Is it possible to make something that could say charge a cell phone, or power a small light?
>>
>>1232711
>This is a rear bike light, not a torch, I want it to be visible from a wide angle
Not much difference, really. Consider this: people who will be approaching you at great speed whom you want to avoid you are most likely going to be, or will have quite recently been, viewing the light from within 15 degrees of centerline, so most of your light should go rearward. Anyone who is outside of that 15 degree cone will be close enough to see you without the light and therefore won't need the same brightness shot in their direction, or has already seen your light from well behind and has given you space. A pair or two of side-facing LEDs wouldn't go amiss, but you'd want to give drivers well behind you most of the benefit, and not waste light illuminating the road behind you. Consider that newer automotive tail lights tend to have spot-like beams for just these reasons.
For the front, a broad cone of light makes more sense, since people turning off of side streets should be able to see you and avoid getting broadsided by you.
>without swamping out the illuminance of street lights on my bike itself
You want to be visible while under the streetlamp, don't you? The inverse square law will take fine care of the drivers' eyes.
>>
>>1232744
I'll cover the inside of the enclosure with Al-foil, so you don't have to worry about excess light going down or up. I'm on the fence about which LEDs to pick, so I think I'll run a test between the white 3mm and piranha LEDs that I bought, and try them with and without the diffuse screen. I'll also put retro-reflective tape on the side of the bike, which should help with people coming from the side.
>>
I'm building a little Pi project using a power tool battery for power. What would you guys suggest using for the contacts?
>>
File: alligator pie.jpg (90KB, 900x897px) Image search: [iqdb] [SauceNao] [Google]
alligator pie.jpg
90KB, 900x897px
>>1232774

Alligator pie, alligator pie,
If I don't get some I think I'm gonna die.
Give away the green grass, give away the sky,
But don't give away my alligator pie.

Alligator stew, alligator stew,
If I don't get some I don't know what I'll do.
Give away my furry hat, give away my shoe,
But don't give away my alligator stew.

Alligator soup, alligator soup,
If I don't get some I think I'm gonna droop.
Give away my hockey stick, give away my hoop,
But don't give away my alligator soup.
>>
>>1232774
Scrap hunt for a broken cordless drill that uses those batteries and take its terminals out. Alternatively, you could form some 0.3mm-1mm copper sheet into the right shape(s) and hold them in place with a housing, probably one made of a solid block of wood, though maybe you could get laminated plywood to work.
>>
>>1232755
Alright, are there any audio guys here who can tell me if a simple common emitter amplifier (made from a TIP31CG) will be able to boost normal 3.5mm jack level audio/a 2.5Vpp triangle wave to get the most out of a crappy 0.5W 8Ω 30mm driver? Or are there better methods?
>>
File: 1487511372115.png (30KB, 852x495px) Image search: [iqdb] [SauceNao] [Google]
1487511372115.png
30KB, 852x495px
>>1232837
Should do, but if fidelity is important, you might consider an LM386, which, being a power op amp in itself, can do some pretty keen stuff on its own. Does Pic related look at all familiar?
>>
File: 1652-03.jpg (135KB, 970x728px) Image search: [iqdb] [SauceNao] [Google]
1652-03.jpg
135KB, 970x728px
Total newb here.
I went and bought an external hiDpi display ( https://www.adafruit.com/product/1652 ) and I'm designing and 3d printing a new anclosure for it.
I feel the external power brick is too bulky and was thinking if I could power the display from a power bank/external battery.
What are my options? Is it possible or viable?
>>
>>1232857
Never seen one, I've never dealt with opamps with more than the default inputs. The project requires 4 opamps (covered by an LM324) and a comparator, for which I was going to just buy a DIP8 dual comparator for and not use one of them, but if I have a use for another opamp then I'll hopefully be able to use a DIP8 dual opamp instead, and I have a few OPA2134 audio opamps lying about which might be what I'm looking for. But it doesn't look like I'll be able to use one of them, judging by its datasheet's lack of thermal characteristics and load impedance values. I guess I'll look about for high-power audio dual DIP8 opamps. Any of pic related look suitable?

I do plan on being able to plug a phone into a 3.5mm socket since I'll be having an amplified speaker anyway, so I guess I do care about fidelity even though it will probably be mono.
>>
>>1232858
Get a tablet-size Lipo cell and appropriate charging and discharging circuits and cram them in the casing, it should barely be another 6mm thickness.
>>
>>1232871
Wait, I'll still probably want to amplify both channels since I'll give installing two speakers a go, so I'll stick to using 1 comparator and whatever dual opamp I can pick, which also means I can stack a couple of OPA2134s if 1 can't handle the power. This circuit is just looking better and better. I'll buy the speakers tomorrow or the day after, so then I'll see what I can do. My circuit is also strictly on a 5V power supply with 2.5V ground rail, meaning a of of the opamps I've been spying won't work because they require 10V between Vin and Vout.
>>
>>1232872
Yeah dude like I said I'm a total newb
Can you elaborate more?
>>
How does one stop being a brainlet?
>>
File: generation-z.jpg (10KB, 300x240px) Image search: [iqdb] [SauceNao] [Google]
generation-z.jpg
10KB, 300x240px
>>1232886
By looking at this image.
>>
File: 1493030587226.png (32KB, 448x558px) Image search: [iqdb] [SauceNao] [Google]
1493030587226.png
32KB, 448x558px
>>1232871
>getting the most out of
Then you want the AN7511. It's got a bridged output and runs off a 5V supply. Just add resistive attenuators and source switching.
The other option would be to add a single op amp and power stage like Pic related, then you can go as big as you like with the output transistors. Cleverly, they use the op-amp's internal output transistors as drivers for the power stage. It probably won't be all that efficient without carefully tuning the bias resistors, but it would work.
>stack
Small-signal op amps are designed to drive rather higher output impedances and lower currents than a speaker requires. You'll spend fifty kiwibux on op amps at that rate.
>>
>>1233021
Those AN7511s are a little on the "I don't want to spend $12 on something that I don't have to", but so is buying another three transistors plus the opamps I already own. I'll check out the fidelity of a CE amp, and if that fails I'll see what the power booster sounds like with BC54s and BC55s at low amplitude as a test.
>>
>>1233032
I thought you had a heap of complementary PNPs and NPNs? If you don't, you should.
I have to ask, anon. What's the context? Is this some sort of Jaycar contest? Is there some qt3.14 at the coffee shop who's offered you an evening of wild sensual pleasures for succeeding in this endeavor? Did you lose a bet?
>>
>>1233063
Nah, just trying to go local for this one. Suplustronics have better prices but worse selection when it comes to opamps. I also have a Jaycar gift card to empty. I got a hundred BC5XXs, 1 TO-220 NPN, 1 TO-220 Darlington PNP (which I'll probably use on a continuity detector or something), a couple of 2N2222s, and a bunch of switches and garbage, along with a couple of opamps. I plan on getting a stockpile of all sorts of components soon, but shipping here takes a month and I've got my mid-semester break in 2 weeks, so I'm in a bit of a bind on that front. If people still care about the wiki someone made, maybe we could make a list of good parts to stockpile there?

I'm making bike lights. By modifying a cheap USB battery bank as the power supply. By putting it in a sardine can enclosure. And I thought I should put a good loud horn on them because I don't have a bell. Then I realised that since I'm going to have loud amplified speakers I might as well put an aux-in socket so I could have a battery bank that's also a portable speaker. So now I need some decent audio quality. Yeah, most of my projects go this way, but at least I'm not installing EL wire anymore.
>>
File: 1489103598184.png (71KB, 787x723px) Image search: [iqdb] [SauceNao] [Google]
1489103598184.png
71KB, 787x723px
>>1233069
Sometimes it's cheaper and easier to separate functions rather than make one circuit to do it all. If only you could separate the horn and amp functions, you could get a separate buzzer module for $5 and easily hook up a switch.
Surplustronics has the TDA2822 hidden in their general ICs section, which is a dual power audio amplifier in a PDIP-8 for NZD4 that can run as stereo or bridged mono, for only 5 kiwibux. For music at a reasonable volume between the hours of 9 and 11, you might be able to get away with the one chip and related passives.
>>
>>1233032
>CE amp
Triangle is not power-friendly. For 0.5W at 8Ω you would need 7.5Vpp at the speaker and a gain of about 3. The gain is not a problem but the 5V limit is. You will end up with an extremely inefficient (10..20%) class A amplifier. Better get a 5V class D H-bridge amp from Alibay :)

Literature: http://whites.sdsmt.edu/classes/ee322/class_notes/322Lecture18.pdf
>>
File: 1423718743283.png (422KB, 665x662px) Image search: [iqdb] [SauceNao] [Google]
1423718743283.png
422KB, 665x662px
>>1230603
I hate these threads, stop spoon feeding the normies. Make them learn the same way we all did
>>
>>1233078
No, you just hate yourself.
>>
>>1233073
I know the buzzers are cheap and they're pretty neat, but since I think it might not be loud enough at only 5V, I decided to use a speaker instead. I already have an oscillator being used for PWM frequencies, so I might as well use that as the horn too, when amplified.

That amp chip looks pretty good, but the lack of a negative voltage rail is confusing me. Is there a reason I shouldn't use this as an inverting amplifier instead of with that stereo circuit, since I'll already have a 2.5V rail for the oscillators. It also says 380mW per channel at 6V 8Ω, so I'll assume 320mW at 5V, which is probably fine for my purpose, unless those missing 180mW from my speakers' maximum are particularly significant. I could stack two.

>>1233075
I could just feed it the square wave instead, for which I guess I should just use a transistor to amplify it, but I'm not sure if that will be very good to the speakers. Hopefully it's not too much of an issue at 5V. For the audio itself those TDA2822 opamps look pretty nice, I'll see if I can pick up one today.
>>
>>1233082
That too
>>
File: outp.png (7KB, 390x186px) Image search: [iqdb] [SauceNao] [Google]
outp.png
7KB, 390x186px
>>1233084
>320mW at 5V
More like 250 but 3dB is barely noticeable.
>>
>>1232723
Possible, yes, practical no. Other electrolytes are far more effective.
>>
>>1232837
TDA2822.

https://surplustronics.co.nz/products/4797-tda2822-2-x-1w-audio-amplifier-ic

Cheap, can run it in bridged mode for moar power.
>>
>>1233137
Already bought one! I'll try it out tonight, hopefully I can get some delicious clipping distortion.
>>
Just got a multimeter. Never done any electronics for real but have watched a bunch of videos and now want to get into it.

Question: which of the components on the circuit board can I probe without soldering them? Someone told me once that you can probe resistors but not capacitors while they're in a circuit. Can someone tell me if this is true and why?
>>
>>1233180
Depends on what you're probing. Measuring voltage is typically pretty safe because you'll have a 1MΩ+ resistor between both leads, so you don't have to worry about shorting anything unless it's above the meter's maximum voltage, or requires very high resistances to work, which is not common. So go ahead and measure voltage across anything that isn't too sensitive, caps, resistors, whatever.

Current is a different beast all together, because it's essentially a dead short though the meter. If you see a voltage source, don't put the meter in current-mode across it. You typically only use current measurement when building a circuit when you have the ability to open up the circuit and put the ammeter in between, because current has to be measured in series, as opposed to voltmeters, which are placed in parallel. The exception to this is if you happen to have a clamp-meter, which can measure current accurately without touching the wire, it instead measures the magnetic field from the current flowing, and you can simply clamp it loosely around a wire.

Resistance measurement has the potential to be worse than current measurement, because it has its own voltage that it puts through components. It measures the current across the other side, and determines the resistance through V=I*R. The general rule is to not put this across a circuit if you don't know what will happen, and it is usually used for measuring seperate components, because it's often faster than calculating resistivity or reading the colour bands. The same applies to capacitance or inductance measurement, and continuity, if your meter has those features, but continuity might be safer.

Typically if you're trying to troubleshoot a PCB, you measure voltages at different points with a multimeter, oscilloscope, and/or logic analyser and see if anything is not as expected, because typically resistances are readable on the board and currents can be calculated manually through V=I*R.
>>
>>1233192
Thanks so much anon!! Much appreciated!
>>
Could someone post the roll for project sheet?
>>
>>1233201
All it takes is a little google...
>>
>>1233204
Thanks a ton anon. Sorry for the request and lack of contribution.
>>
>>1233204
Roll
>>
>>1233204
Too bad /diy/ is too slow to get enough RNG. This ain't a roll because I've got enough on my plate. You should just pop into a /b/ thread and make a generic banana-tier post to roll instead. Or just pop on over to an RNG site.
>>
Hey /ohm/, any of you take the FE? I take it Saturday and haven't really studied for it. Looking at the subject matter, the only thing that I think will need review is transistors. But even that I feel cocky enough about to not worry. I've never taken Controls, though, so more or less chopping that up as a loss.

The "handbook" being so remedial, and from stories I hear from other EEs, makes me think it's going to be a cake walk. Am I in for a rude awakening?
>>
>>1233084
Ok, I might see why you don't use high power opamps off a normal ground rail, because the power drained through this rail is very significant, and therefore also significant through whatever linear regulator, in this case another opamp, creates this voltage rail. So I don't understand that circuit (Figure 1), but I assume it relies on C2 to treat the inverting input like some sort of ground? I'll try it anyway since it is a recommended circuit for powering speakers and such, but I'll run it cautious with a gain pot and volume pot.
>tfw it says "Obsolete Product(s)" all over my datasheet
>>
File: 1473762156685.png (59KB, 780x474px) Image search: [iqdb] [SauceNao] [Google]
1473762156685.png
59KB, 780x474px
>>1233241
>why you don't use high power opamps off a normal ground rail,
You can, and often you do, when you have a proper split supply. This device was optimized for single-supply usage.
Amazingly, manufacturers other than ST give internal schematics in their datasheets, see Pic related. The red is the "normal" built-in feedback path. The blue is the feed from the top unit (in-phase) output to the inverting input of the bottom unit. I suppose C2 just helps keep C1 polarized and the resistors are chosen to provide -1 gain in bridge mode.
>>
>>1233267
Whoops, move the red line over to the other side from OUT1 to the resistor to the base rather than OUT2.
>>
>>1233267
Built-in feedback, that's pretty clever. I'm testing atm, will report back soon.
>>
>>1233283
Alright, pretty peeved because one of the shitty speakers was open-circuit on arrival, but that might have been just overcooking it on the iron. Shoulda' tested it first. But I'm getting great audio quality on the low volume end, though at the high end I'm getting a great deal of interference, probably of the clipping variety as it seems stronger when the song gets louder. It seems I can't get it too loud without hitting the rails of the opamp. But on the bright side, the distorted music sounds dank A.F. and after 10 minutes of annoying my flatmates at 11:30 the thing isn't even warm, though it is only on one channel. I found I can't adjust the volume by replacing the resistors R1 and R2 with 1k potentiometers, though I might have done that a bit wrong. Anyway, I can adjust the volume well enough on my phone itself.

Now I think the too-low output voltage the sort of problem that can be solved with an audio transformer, but chances are any ones I'm likely to fins in a store are going to be a little pricy, so how many turns would I need to make a decent one myself? I'm thinking 1:2 ratio, but maybe up to 1:4. Inductance in the range of maybe 10-50mH, depending on patience. Perhaps the "impedance" of the transformer coil(s) compared to that of the 8Ω speaker be a useful comparison to make in this instance.
>>
Oh shit, there's a deal coming to Auckland NZ, and it's in the form of a bunch of shitty refurbished/ex-repaired stock that salvage monkeys like myself adore. I hope they have stock and prices like EEV-blog Dave saw at the Australia warehouse sale, because if so you'll be able to pick up everything for 90% off if it doesn't sell.
>>
So I ordered some chips in DIP format, they came SOIC, which I've never used before. But they took 6 weeks to ship so I'm not bothered returning them, I can make it work. However: how do I figure out which way is up for the pinout? The datasheet diagram has this circle on the dimensions page, but the chips have no circle, and the pinout has a little inverse nubbin on the top end, but the chips have no nubbin. They have a triangle, which is in the bottom left corner if I orient it so that the text is the right way up and the triangle points upwards, which is the right way up for a triangle for any reasonable person.
>>
File: Capture.png (8KB, 286x282px) Image search: [iqdb] [SauceNao] [Google]
Capture.png
8KB, 286x282px
>>1233323
>>
>>1233323
there should be a circle or indentation

but if there isn't, the side with the bevel sloping from the top surface to the pins has pin 1

make it face you and the pin on your left is pin 1
>>
>>1233329
The top is featureless excpet for the text, no indentations at all

One side does seem to slope up at a less steep angle than the other though, is that the bevel you're talking about?
>>
>>1233338
Big fucking notch on the right hand picture isn't a clue?
>>
File: marking.jpg (12KB, 508x226px) Image search: [iqdb] [SauceNao] [Google]
marking.jpg
12KB, 508x226px
>>1233338
which chip is it?
>>
>>1233345

It's a lot more obvious in the picture than in real life and it varies what side it's on from chip to chip. I figure it's a manufacturing mark, like where the plastic's snapped off a die.
>>
>>1233377
ATtiny
>>
>>1233377
I drew on the triangle BTW, to show its apparent position if you use the text to indicate which way is up. That's not in the datasheet, just the nonexistent circle.
>>
>>1233377

as the other guy said, some show it by a bevel
>>
File: Untitled.png (22KB, 944x640px) Image search: [iqdb] [SauceNao] [Google]
Untitled.png
22KB, 944x640px
Say I've got a basic LED control circuit with the ground of the LED going through a transistor that I toggle. Is there an advantage to having the control voltage on the base and tying the emitter to ground, so that with a positive voltage opens the transistor and turns on the LED, versus tying the base to positive voltage and applying control voltage to the emitter, so that turning on the control voltage opposes the BE voltage/current, turns off the transistor and so the LED? I've seen the latter a few times but it would never occur to me to do it that way.
>>
>>1233414
Right hand side not meant to be grounded, soz
>>
>>1233414
I'd apply voltage to the base and use a resistor to keep the base current under 10 amps.
>>
File: bevel.jpg (6KB, 262x200px) Image search: [iqdb] [SauceNao] [Google]
bevel.jpg
6KB, 262x200px
>>1233382
>I drew on the triangle
Yes, I should have seen that it was a doctored drawing and probably the wrong outline (8S2 EIAJ). Your pic looks more like S8S1 JEDEC, non-square body. Solved I think. Next step would have been the multimeter..
>>
I'd like to join an electronics club. Trouble is, my university doesn't have one. Now, I would just start one, but what's making me hesitate is the fact that I don't really know much about electronics beyond the theory I studied in class. I feel like if I am to be the founder (and president?) of my university's electronics club I should be in charge of giving workshops or helping out the newbies, which of course I can't do because i'm a newbie myself.

What do the good people of /ohm/ think? Perhaps I could take a purely administrative role and let someone more expert take care of the technical side? Also, I'm not even sure what activities the club would do. I'd just like it to be something where nerds can meet and chill while showing each other's projects and helping each other out, but perhaps that might seem too boring to attract numbers? I feel like it would depend too much on people doing a lot of side-projects to keep the club interesting...
>>
>>1233414
>Is there an advantage..
Emitter is in the load path and carries the LED current, base current (control path) is only about 1% of that.
>>1233420
>base current under 10 amps
Excellent idea.
>>
>>1233443

Find a partner. There's always someone in an engineering class who gets intuitively how things work better than anyone else. Find that person and see if they're up for it.
>>
>>1233443

dont do it. people expect the club founder to be especially competent and interested. when they find out you're a noob, they'll just figure you just wanna meet guys coz yr a phaget.
>>
File: IMG_1134.jpg (2MB, 3264x2448px) Image search: [iqdb] [SauceNao] [Google]
IMG_1134.jpg
2MB, 3264x2448px
I got the Jaycar $5 shill bag, and what kind of joke is this book? They had a little old-new and repaired stock, but nothing was hugely cheap except for a $200 pair of speakers down at $19, which were of no use to me. The rest of the stuff on sale was all outdated home brand wifi gadgets and shitty security stuff, so scavenging parts out of those would require becoming an I2C master.

I really only got the bag because it gives me something to put cold groceries in, but the torches look ok.
>>
How low a voltage drop can you get across a constant current source circuit? I'm looking for 40-50mA at ~2.4V out of two 1.5V cells, is this possible?
>>
>>1233462
I know two guys like this at my school: one stopped right after his bachelor's because he already got a job, the other is probably too autistic to accept any sort of leadership role (which is a shame because he's very smart).

>>1233514
Main reason why I'd like to join a club like this would be to learn more in a group, because in my major we don't learn as much about electronics as i'd like since it's more about high power installations (three-phase systems, generators, etc.). Then the idea of founding a club came because at my uni there's fuck all to do after classes. Here in Europe clubs aren't really a thing like in the anglo world, so once the last class of the day is over everyone just scuttles away to their homes. I'm really jelly of American universities (except for the tuition fees).


(subscribe to my blog etc.)
>>
So I'm trying to calculate the insertion error generated by a multimeter in a current measurement. For that I'd need the internal resistance value for the given scale right?
But the multimeter manual gives me only a tension drop rating for each DC current scale. How do I use that? I've already solved the mesh equations including the multimeter taking it's resistance as another variable. Pls help i'm a brainlet
>>
>>1233863
>tension drop rating for each DC current scale

if that means voltage and it's constant then that is one answer. the power consumed would of course be the current times that voltage.

or you could use another multimeter to measure the voltage across the current meter. that adds another very high impedance to your circuit.
>>
>>1233868
Sorry I'm a retard, instead of using a resistance I just add the voltage drop when doing the meshes.
>>
File: 1LED-1AA.jpg (4KB, 225x169px) Image search: [iqdb] [SauceNao] [Google]
1LED-1AA.jpg
4KB, 225x169px
>>1233691
>the torches look ok
Do they run on a single cell? If 3xAAA it's the usual inefficient crap.
>>1233799
>across a constant current source circuit
Across the complete circuit, i.e. a 2-pole cc source? That's a challenge. A 4-pole solution should be possible but it would be easier with a single lipo. 1.5V is soon 1.2V.
>>
>>1233873
Really garbage, I thought the outside was silicone at first glance but it was just that hard plastic with a somehow soft feeling to it. 9 LEDs in parallel, but no dropper resistors. I measured 3.8V across the LEDs, sagged down from 4.5V. I guess they chose LEDs with particularly high forward voltages.
>>
>>1233863
The inserted resistance is constant for a each current range, tension drop/full scale current.

In general the resistance is full scale current * full scale voltage at the most sensitive DC voltage range. This becomes obvious when you realize how an electronic meter measures current.

My 4000 count MM has only a single 400mA DC range and a DC sensitivity of 400mV, both full scale. So the resistance of the mA range should be 1Ω but it is 1.4Ω because it has a fuse and a (rotary) range switch. The switch ages and will become a problem after a while. I almost never use the current range but insert a small (like 0.1Ω) resistor into the current path and measure the drop across it.

For some reason the tension drop you mention is sometimes called 'burden voltage'.
>>
>>1233887
The manual only gives these things regarding DC_A
>scale
>resolution
>precision
>voltage drop (0.2V for all scales)
>overload protection
How would I go to find the internal resistance?
>>
>>1233893
>>1233887
In the DC-V part it learly says
>for this scale the internal resistance is X
>>
Is it ok to buy oscilloscopes from alibaba, or should I insure my house against fires beforehand?
What's a good babby's first scope?

What about lab power supplies?
>>
File: fp6kj00[1].png (4KB, 407x169px) Image search: [iqdb] [SauceNao] [Google]
fp6kj00[1].png
4KB, 407x169px
>>1233898
>>1233893
It's just an exercise from my circuit book, I decided to build it and test it so see if the math was working.
>>
>>1233893
>voltage drop (0.2V for all scales)
Sure, because it's a 200mV meter. But the effective resistance is 200mV/current range
>>1233898
>for this scale the internal resistance is X
Oh, is it an analog meter?
>>
File: f100_no_lights_on2.png (43KB, 800x480px) Image search: [iqdb] [SauceNao] [Google]
f100_no_lights_on2.png
43KB, 800x480px
>>1233899
>What's a good babby's first scope?

you can get a usb scope that is fine as long as you don't screw up and kill your laptop. or if you have $400 the Rigol DS1054Z is a great hobbyist scope.
>>
>>1233904
The amperimeter in the pic is digital, it has scales. Like 2mA, 20mA etc. The voltimeter is a cheap chink analog I found that has the internal resistances in the manual.
>>
>>1233906
thanks. I'd rather not mess with my laptop.
>Rigol
How does it compare to this one? http://www.befr.ebay.be/itm/OWON-SDS5032E-digital-Storage-Oscilloscope-SDS5032-Scope-Oszilloskop-/131824586195?hash=item1eb15b8dd3:g:oSQAAOSwBLlVLBI0
>>
>>1233899
Don't listen to that other guy, everyone here finds their oscilloscopes in the dumpsters of engineering firms. Bonus points if they're from the 80's.
>>
>>1233912
>Bonus points if they're from the 80's
Does that mean that I have to grow a pedo stache and a mullet before getting mine?
>>
>>1233911
Owon's are crap, you don't want to cheap out just a little bit and find yourself regretting your decision. Rigols have a pretty good reputation.
>>
>>1233911

lol I don't even recognize that language much less read it. from the picture it looks like it has 3 channels (or maybe just 2 with some other kind of input); the rigol has 4. other than that it "looks" like a good scope. I'd read some online comparisons if you can find them. I know there's lots of discussion of the Rigol specifications.

One thing you will read about is the "free upgrade" which doubles your bandwidth, ram, and adds other features. I don't know if that still works and I don't plan to try it because the $400 dollar model is pretty awesome as is.
>>
>>1233914
>>1233916
Thanks for the advice. I guess i'll go for the Rigol.
>>
>>1233907
Then it all makes sense. At the core, analog (magnetic) meters measure current, digital (electronic) meters measure voltage (tension). Both have shunt (parallel) resistors for current. The electronic meter has a divider chain for voltages and the magnetic meter has a chain of series resistors.
>>
Posting here to make sure I'm not crazy or stupid before I go over to the Microchip forums.

I wrote a function that uses assembly code to control WS2812B RGB LEDs. It can be found at https://pastebin.com/b0AgJYAK (embed)
This is code for a PIC16F716, so you can look at the instructions in the datasheet at http://ww1.microchip.com/downloads/en/DeviceDoc/41206B.pdf

So what's this dummy variable, you ask? I have no idea. If I don't have it, the red LED PWM byte gets copied onto the green LED variable. I did some testing with a modified function that only moves [email protected] to a register, then does the per-bit output pulse width modulation. I found that that the value in the register is still [email protected], even if red is never used in the function.

So I added in that dummy variable in the function declaration and never used it, and the function works fine.

Am I missing something, or is the compiler broken?
>>
>>1234003
>what's this dummy variable
game the compiler. add unsigned char voodoo to the list and see what happens. if it fails again the compiler may not like an odd var count or not less than four. in this case add another to check. only the secret-ary knows.
>>
I'm an electric retard here, so help me out.

I'm trying to retrieve files from my laptop's crapped out hard drive. It's not infected with a virus or anything like that, it just failed. Can I just hook it up to my desktop's SATA or is there any preparation I need to do?
>>
>>1234030
I used a sata to usb adapter and it worked.
>>
File: corned_beef_48342.jpg (42KB, 564x423px) Image search: [iqdb] [SauceNao] [Google]
corned_beef_48342.jpg
42KB, 564x423px
>>1234030

if it has SATA connectors then it does not care what sort of thing it is connected to.

"not a virus, it just failed."

seriously, are you from redit? how fucking hard is it to plug in a flash drive and backup your shit? are you mentally retarded? plug in an SD card or a flash drive and back the shit up the moment you save an important file.

nothing pisses me off more than "OMG MY LAPTOP DIED AND MY FILES ARE GONERS"

well, stupid is as stupid does. kiss your files goodbye.
>>
>>1234083

It's not important files, those are backed up. It's just my porn. Chill your autism out.
>>
File: 5glwaXdHc4E.jpg (29KB, 640x425px) Image search: [iqdb] [SauceNao] [Google]
5glwaXdHc4E.jpg
29KB, 640x425px
>>1234086
>It's not important files,

so why are you here, again?
>>
>>1234090

Advice, how bout you?
>>
This thing arrived a few days ago; so far it's alright, apart from not holding calibration after turning it off, but I got the replacement NVRAM chip from Maxim as a sample. The resolution is really low like one anon mentioned, I'm thinking of selling it again as working and getting something else.
Could I get 300 bucks for it in that condition with all the accessories, or is that wishful thinking?
>>
File: nat_prety_423234.jpg (207KB, 480x640px) Image search: [iqdb] [SauceNao] [Google]
nat_prety_423234.jpg
207KB, 480x640px
>>1234098

so you bought a piece of shit and you want to pawn it off on some other trusting anon?

few days ago. got nvram chip as a sample

why can't you buy stuff, anon?

decency. it's old-fashioned, I suppose.
>>
>>1234101
I was probably gonna sell it on ebay, unless someone here wanted it.
Why pay for something when you can get it for free? I don't have a lot of money to spend on this hobby, saving $30 is a big deal for me.
Why are you being such a dick?
>>
>>1234101
>>1234090
>>1234083
Imagine being this guy.

I hope you find happiness, anon.
>>
>>1234101
>>1234090
>>1234083

The salt is real
>>
>>1234104
>Why are you being such a dick?

I apologize. You don't really suck.
>>
File: nice_man.png (273KB, 586x346px) Image search: [iqdb] [SauceNao] [Google]
nice_man.png
273KB, 586x346px
>>1234105
>>1234106

samefag is not nice
>>
Does the assembly instruction MOV clear the register you're moving the data from, or is it more of a copy instruction?
>>
I have a 16.5V AC plug in transformer from a security system. Is there a way to use it to power basic components on my breadboard without an AC/DC converter?
>>
>>1234149

MOV does not clear the source register. it leaves it intact.
>>
File: file.png (63KB, 399x296px) Image search: [iqdb] [SauceNao] [Google]
file.png
63KB, 399x296px
>>1234149
You'll have to do some math to figure out the size of the cap you need based on the noise you can tolerate.
>>
>>1234157

no, you need at least a diode and a capacitor to rectify and filter, then probably a voltage regulator to cut the voltage down to useable size.
>>
>>1233913
no, clean shaven and flat top

>>1234003
are you clobbering the stack?

>>1234092
>>>/g/

>>1234149
copy
>>
I'm tired of having to just make due on my little projects instead of having the proper tools I need, so I'm going through and spending $100-$200 on just getting the little optional extras to make things run smoother. Engineer solder pump, bending guide, round nose pliers, a handful of breakout boards/headers, color coded sets of wire instead of whatever random spool I have left, etc. For stuff where you're working with 22AWG wire, do you stick with just solid, stranded, or do you use both? I have solid in my cart, but don't know if I should get the silicone insulated stranded instead/as well.
>>
>>1234212
I'd go stranded silicone + regular solid core. You don't really want to use stranded to wire together a veroboard or similar, you don't want to use solid core for external connections.
>>
Gore warning.

So I fucked up and due to excessive re-soldering the soldering pad has come off completely. What's the best way to fix/replace it?
>>
>>1234262
>the best way
is not to expect a 'best way'. Remove the gunk and see what has survived.
>>
>>1234272
it looks pretty bad, but the area still works, there's just no longer any point to solder the wire onto. I'm asking what I can do to replace the soldering pad.
>>
>>1234274
What do you mean gone? post another picture.
Worst case, sand off the solder mask and attach there.
>>
>>1234278
Sorry if I'm being unclear. The copper pad underneath is still there and still works, but the circular soldering point that was above it has come off.

Would I be able to solder directly to the copper?
>>
>>1234281
And apologies for the blurry photos, I can't get my camera to focus on it properly.
>>
>>1234281
If it was just a little silver bump, that's just solder, you can just add more yourself. If you mean there was some sort of metal wire terminal like a hook or a loop you should probably replace it. It should still be fine if you can find it and clean it off.

To make it clearer, solder is a metal SOLVENT, which means it dissolves metals like copper just like hot water dissolves salt. So when you're soldering a joint, you're melting the wire and the pad, and then holding them together until they cool and solidify. it's like low-temperature welding.

But, because of the way it works, ANYTHING between the two metal surfaces will make it either fail, or form a poor connection. You've really got to scrape that area clean and flux it good before trying again.
>>
>>1234291
> solder is a metal SOLVENT, which means it dissolves metals like copper just like hot water dissolves salt. So when you're soldering a joint, you're melting the wire and the pad

This isn't how soldering works. The melting point of copper is ~1000 degC, which is much higher than any soldering iron.

Pads get damaged by delamination, where soldering melts the resin holding the copper to the rest of the board, allowing the pad to move and be torn away from the track.

It's possible to repair this, but not easy. Particularly if your "handiwork" isn't good (and delaminating a pad suggests that).
>>
>>1234366
he says the copper pad is still there, and honestly it's so fucked up I can't tell what I'm looking at.
>>
>>1234281
>jpg
>241x402
>blurry
>2017 CE
how
>>
File: gone.jpg (4KB, 159x158px) Image search: [iqdb] [SauceNao] [Google]
gone.jpg
4KB, 159x158px
>>1234291
>melting the wire and the pad
The solder doesn't melt the copper, it's a diffusion process where the solder becomes an integral part of the molecular structure of the copper surface forming an alloy. You can no longer separate them by peeling off the solder from the copper. The same process works with silver or gold but not e.g. with stainless steel. The brown area is definitely not copper, it's the base material of the board where the copper trace has been delaminated and removed. The damage can only be repaired by reconstructing the lost traces by means of thin copper wires. This will require skill and good tools.
>>
File: QJ23a8T[1].png (145KB, 881x375px) Image search: [iqdb] [SauceNao] [Google]
QJ23a8T[1].png
145KB, 881x375px
Can somebody help with this circuit? I need to find the power in the controled source. It's obvious that it's (Alpha)V0*If. But I can't reach the value of V0 using either meshes or nodal analysis. send help
>>
File: midi_keyboard.jpg (73KB, 1200x1200px) Image search: [iqdb] [SauceNao] [Google]
midi_keyboard.jpg
73KB, 1200x1200px
What makes MIDI keys work? What type of 'button' is used to make the keys feel soft when you press them lightly instead of harsh (like regular buttons)? How is the velocity calculated based on how hard you press a key? Where can I order these individually without having to buy a MIDI controller myself?
>>
>>1234421
How would you do it if you had to?
Two contacts instead of one? Time?
What is easier to move when slow?
>>
>>1234385
have you tried with supernodes or superposition? Either way I'm not sure how to solve it either, that vertical wire second to the left throws me off. i0 should be zero and the current source is shorted...
>>
>>1234432
>second to the left
meant second from the right
>>
>>1234434
>>1234432
Thanks anon, all the circuit professor in my uni are cray cray, so I had to chose the one that at least tries being good and gives his old tests online. (They are hard :( )
>>
File: sort.png (4KB, 663x176px) Image search: [iqdb] [SauceNao] [Google]
sort.png
4KB, 663x176px
keep /ohm/ afloat
>>
>>1234569
new thread when?
>>>/g/61928360
>>
>>1234421
A pair of boring rubber dome switches set to activate at different points while pressing down the key, the time difference between when the first touches until the second touches is used for the velocity.
>>
>>1234569
floating inputs are bad

>>1234576
how bout now?

NEW BREAD
>>1234666
>>1234666
>>1234666
NEW BREAD
>>
>>1234098
You paid $100 for it, apparently expecting it to be in working condition. At least you didn't mention it was broken earlier. Yet, after fixing it, you don't think it is worth the $100 you spent on it.
Why the hell would anyone pay $300 for it? You can get a new, more useful scope for that money.
Thread posts: 330
Thread images: 73


[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y] [Search | Top | Home]
Please support this website by donating Bitcoins to 16mKtbZiwW52BLkibtCr8jUg2KVUMTxVQ5
If a post contains copyrighted or illegal content, please click on that post's [Report] button and fill out a post removal request
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.
This is a 4chan archive - all of the content originated from that site. This means that 4Archive shows an archive of their content. If you need information for a Poster - contact them.