/ohm/ - electronics general, many midi meechi edition

File: key-contacts-bottom.jpg (872KB, 1600x1200px) Image search: [Google]

872KB, 1600x1200px

>>1234421

another way is to have several concentric switch contacts, so that the harder you push the button, the more contacts are made.

couldnt find a pic so you'll have to imagine concentric rather than side-by-side as in my pic.

File: SSS-02-800x800.jpg (117KB, 800x800px) Image search: [Google]

117KB, 800x800px

You're putting together the best setup you can for your electronics hobby. What are the gadgets and products you get?

Pic related, Engineer brand solder pump.

File: 8xQFV.jpg (204KB, 1625x1080px) Image search: [Google]

204KB, 1625x1080px

>>1234678
Another way to do variable pressure is a resistive pad on the board + a dome shaped contact on the button, as used on Playstation 2+ controllers

>>1234685
instrumentation aside, i'd get a reflow toaster oven, a 3d printer, and a three axis platform that i can mount a vacuum needle to for pick and place. it's really hard to solder qfns when you don't even put them down aligned right in the first place. come to think of it i can afford these things but i'm too jewish to buy anything above $60.

>>1234701
>it's really hard to solder qfns when you don't even put them down aligned right in the first place
$30 USB microscopes to the rescue?

>>1234717
i have one. i think i just need more practice since i've only done a couple of them, but i'd rather just automate it.

File: keyboard enconder.png (146KB, 809x1010px) Image search: [Google]

146KB, 809x1010px

Anyone know of a way cheaper version of one of these keyboard encoders?

I've seen like 20$ arcade style encoders, but I don't know if those handle the keyboard "matrices"/ all keys on a normal size keyboard.

File: 996-main-500x500.jpg (24KB, 500x500px) Image search: [Google]

24KB, 500x500px

I've seen this motors everywhere, they seem big, but they're actually 12mm wide with a 3mm shaft.
Is there a version of this motor (or similar) with a 5mm shaft?

>>1234666
Wtf is going on in this picture? Why are they burning a breadboarded IC with a soldering iron? Why is the scope attached if they're still soldering? Why is there a torque wrench on the desk?

>>1234779
Stock pictures. "We need a DIY picture, get stuff labeled DIY from the prop box and put it on the table."

>>1234781
That picture was some anon making fun of stock images.

>>1234786
Also note that, as with previous OP pics, the soldering iron is being held by the element (which gets hot when the iron is powered).

File: 1485119083865.jpg (30KB, 650x609px) Image search: [Google]

30KB, 650x609px

>>1234719
Soldering with paste is pretty forgiving, in my experience, as long as you're at least half on the pad.

File: probe-fake-fume.jpg (45KB, 502x673px) Image search: [Google]

45KB, 502x673px

>>1234666
Great image, the fake fume is back!
>>1234779
>Wtf is going on in this picture?
Things designed to derail you. The probe is attached to the calibration output of the scope, but all inputs are open. The screen shows a stored image, before adjustment.

>>1234806

Isn't that just the probe not being adjusted?

>>1234807
Probably, but you can't adjust it unless connected to an input.

>>1234816

Shit, now I see it. Derp.

>>1234816

maybe it's a wireless probe so you don't have to use a dangerous isolation pluge.

That's how you read out the contents of a protected micro.

File: $_57.jpg (24KB, 900x738px) Image search: [Google]

24KB, 900x738px

I need a way to detect force applied to a flat disk, about the size of a CD.

Basically any kind of electrical indication that something is placed on the disk will do. The amount of force is not really important.

I was looking at pressure sensitive resistors, but they're all pretty small and expensive.

>>1234850
Too vague. Sensitivity, displacement, environment, drawing? Did you notice that someone already posted one possible way on this page?

>>1234850
two thin conductive disks separated by a thin insulator ring around the edge, when you place an object on them it presses them together. Post prototype this week thanks.

https://www.edx.org/course/introduction-electrical-electronic-tokyotechx-ee101jx
>2017
>not learning about electronics with your waifu
what's your excuse?

File: 20170827_222207.jpg (1MB, 2560x1536px) Image search: [Google]

1MB, 2560x1536px

>>1234666
Ive not posted in any of these threads, but i have a relevant question... I'm looking at a solid state guitar amp that's buzzing, etc.. i took out the board and found this, these are obviously bad?

>>1234860
conductive foam (the black stuff ICs are packaged in) can work

>>1234916
bad? just add some solder

File: 2017-08-28 12.43.25.jpg (264KB, 654x403px) Image search: [Google]

264KB, 654x403px

>>1234916
.cont

The question is i found this on the other side, elsewhere on the board. This is not normal? Or did they do shit like this on purpose to connect the resistors or something? If it's another problem I'd like to take care of it now, but I'm unsure because I've found a bunch of cold joints which are obvious problems.

>>1234920
Ok, thanks. I was just gonna try to reheat it but I'll add a little more.

When people throw electronics in the toaster oven as a janky diy reflow oven, how does it not melt the plastic on the connectors and burn out parts? Are most components okay at solder temps if they aren't powered? And do they use some high temp plastic or something? I know the fix for the Xbox 360 being shit was to put just the board in the oven, connectors and allm

>>1234982
How do you think the board was soldered at the factory?

>>1234779
how the fuck are you going to torque your electrolytics to 10-12 ft-lbs without a torque wrench?

File: temp_profile_2.png (72KB, 666x632px) Image search: [Google]

72KB, 666x632px

>>1234989
Rows and rows of toaster ovens. When it smells right you pull it out and throw the next one in.

>>1234765
Any microcontroller with usb hid drivers available
Or if you just want it to be easy and cheap, an arduino leo and some piso shift registers.
There's libraries with nkro support too. Though that might go against the spirit of this thread.

>>1234921
Looks like a bodge to correct an error in the board design, it's fairly common for that sort of equipment.

>>1234778
there are larger ones on banggood, ebay, and similar websites.

pololu also sells the same thing so you can find more technical specs on the pololu motors and see if the chinese match up

>>1234765
The Teensy development boards are pretty popular for DIY keyboard controller replacements.

>>1234779
Reminds me of the sandnigger kid with the clock.

If I'm programming an 8-bit MCU in Arduinpo IDE and I want to deal with, say, the number 10,000,000, will the compiler do magic things to make it work or would I have to do something like nested while(i<=255) loops? In Assembly I'd just use ADC and accumulate it over time but I specifically have to use Arduino.

>>1235059
arduino can handle uint32s

>>1235066
Yeah but I mean if I'm programming in the Arduino IDE for like an 8-bit AVR, just using an Arduino as an ISP, not actually writing code to run on an Arduino, is there something in the compiler that will figure out something has to be done to break up a large number calculation for an 8-bit ALU

>>1235071
Arduino itself is literally an 8-bit AVR with a serial bootloader

>>1235071
the atmega328p that most arduinos use (at least the ones i have) does not have a 32 bit ALU so the compiler must use 32 bit variables to abstract multiple 8 bit operations. whether that means the compiler will work in your case is beyond me since i'm not a compiler.

>>1234921
Nah, that's standard Crate rework.

>>1234982
Temperature-sensitive components are installed after reflow. Liquidus temperature is reached only briefly, long enough to reflow the solder paste. And connectors are often made of high-temperature plastic or thermoset epoxy.

>>1235071
>is there something in the compiler that will figure out
Yes. If the compiler supports uint32_t then the compiler will produce code to provide the defined behavior for uint32_t. You could always just try it and disassemble the result with objdump.

>>1235021
>>1235052
Thanks

>>1235080
>>1235078
>>1235076

Neat, thanks bros

>>1234666
For god sake, op, compensate this probe

>>1235102
a little overshoot helps keep you on your toes

>>1234666
In all the times I've seen this image this is the first time I noticed the socket wrench

File: 1.jpg (25KB, 600x480px) Image search: [Google]

25KB, 600x480px

Image for the next thread

File: 1421156812881.jpg (7KB, 225x224px) Image search: [Google]

7KB, 225x224px

>>1235116

>>1235106
Also the sharpie.

>>1235116
Is... is that a motherboard just hovering in the background?

>>1235139
It is sitting on a white/beige case.

What do you call the phenomenon where a floating node's voltage changes in response to its capacitance with your body as you approach it? Capacitive coupling?

>>1235080
>Crate
Nailed it. Guess you've seen this before

>>1235139
>is that a motherboard just hovering
obviously it's a hoverboard

>>1235165
SLM was the giveaway. I had one of those things back in the day. Sight unseen, if buzzing, you could try adding a bit of input load (add 1M resistors across the input until the desired effect is reached) or increasing the filtering caps. Also check signal cable routing (if any) internally and ensure that the shields are grounded.

>>1235170
Thanks. Hopefully i can get it to quit buzzing enough to pawn it and get something for it. It's a gfx 2x12, says 1998 on the board. I got it like +10 years ago for 50$. Wasn't buzzing then, though

>>1235224
>almost 20 years
Oh, you'll probably want to change out the power supply filter caps first.

>>1235167
kek

>>1235239

I've had one that had a seal between the tip and the reservoir that needed to be broken, first.

I've also had one that I thought was fucked, but just need an excessive amount of pressure to actually get flux out of the reservoir into the tip.

>>1235230
> pointing out 20 years
Damn i feel old now... but yeah, the cold joints in the first pic are actually to one of those caps.

Where can I order cheaper brand electrolytic capacitors that are still sourced first hand from their suppliers? Digi and mouser only have exorbitant nichicons and the like for higher voltages.

>>1235292
Digi-Key and Mouser generally distribute only authorized product. UCC and Cornell-Dubilier are ok-tier. How big did you need?

>>1235293
A bank of 350uF rated 500V+. The value of any individual capacitor doesn't matter as long as they're less than 1.5" tall. I'm looking at $20 on digikey and wondering what a worse brand like capxon would cost.

Is it possible to build a DIY oven stabilized reference or oscillator or would you pretty much have to get this manufactured somewhere?

This isn't a for a particular project, I'm just wondering in general for future reference.

>>1235299
There's no reason you can't thermally couple your reference, a resistive element, and a thermocouple and then insulate that whole assembly. There's sources of drift besides temperature though and you'd need a better reference to calibrate it to.

>>1235299
Sure, you build a temperature-controlled oven and you put an oscillator in it. Cheeky breeky! Tuning it is another matter.

>>1235298
Oh, for a 500V cap $20 is about right. How about two 680uF/250V+ in series for under $10?

>>1235305
The series option is way cheaper than I expected it to be. A whitepaper tells me I shouldn't waste more than a watt in balancing either.

File: DC-Motor-RS-775-.jpg (75KB, 633x569px) Image search: [Google]

75KB, 633x569px

Noob here.

Can you increase the torque on a DC motor by strapping 9v's together?

I'm trying to get a mechanism of constant speed for an experiment, but I'm not sure the motor I have is going to overcome the mechanical resistance, if you raise the voltage, will the torque rise with it?

>>1235348

yes. the primary risk is overheating the motor.

>>1234646
>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.
thanks

>>1235348
Torque is related to current, speed is related to voltage. Wiring two 9V blocks in series doubles the internal resistance, connecting them in parallel halves the internal resistance. If you want more torque you need more current. 9V blocks are the most inefficent batteries around and should only be used to power low current electronics, multimeters and the like.

Simple experiment: Connect a multimeter to a 9V battery and note the no-load voltage. Additionally connect the motor and note the change. Grab the shaft and apply torque until stand-still. Note the voltage. This is the effect of the internal resistance of the battery.

>>1235369
If I remember right every key has its note number and sends noteoff when idle by holding the idle contact closed. When pressed it sends noteon by closing the noteon contact and the time between the last noteoff and the noteon is the velocity. When you hold it down and change the pressure, the aftertouch information can be evaluated to modify properties of the playing note. You could use that to emulate the sustain pedal of a piano. And then there's the global pitch wheel.. The polyphonic capability is only limited by the controller.

>>1235377
This means 'idle' should better be called keyup, which is an active state that continually sends noteoff to prevent 'hanging' notes while keydown sends the noteon message. The noteoff message is ignored be the controller if this note is not playing.

>>1235387
No controller I know of sends note off continuously. Just once is enough, 99.999% of the time. The All Notes Off message exists for that 0.001%.

>>1234989
Hadn't really thought about it before, but mostly wave soldering I figured.
>>1235080
What kinds of things are usually temperature sensitive enough that janky toaster oven reflows are a bad idea?

Noob here, i've only ever done electronics on paper:
How important is it to insulate the underside of a perf board, particularly when building a power supply or other HV circuits? I suppose it is normal to use standoffs to attach the board inside a case, but is that enough precaution? Should there be a protective earth cable connected to the inside of the case?

>>1235547
Poke a component with a soldering iron, if it's not melting then it's probably fine. Most boards are soldered in temperature controlled ovens, and it's not that hard to make a toaster oven temperature controlled.

I've only ever done little repairs, but having my only flux be in the middle of a piece of solder is kind of shit, so I'm grabbing a flux pen. Are they all created equal, or should I avoid/aim for certain ones?

>>1235552
grounding metal cases exposed to any voltage is important but that's a human safety feature, it won't stop lead-case-lead shorts from destroying your project. standoffs are your best bet for that, hence why many electronics cases have them built in.

>>1235555
>lead-case-lead shorts
just to be clear, you mean when the mains voltage is transmitted directly to what was supposed to be the LV DC lead attached to my project? True, I hadn't even thought about that.

Also, check'd

>>1235547
If your reflow toaster oven heats roughly evenly, and follows the profile closely enough, there should be no problem. Electromechanical parts like switches and connectors are usually the ones to watch for, followed by electrolytic caps which will be okay if you follow the profile strictly. Large ICs should probably be dried before baking.

>>1235554
RMA flux is stronger, i.e. cuts through oxides better, but requires thorough cleaning after soldering. No-clean flux is weaker but doesn't require cleaning. Which you would use depends on the age of your components and boards, components' sensitivity to water or IPA, and whether you can easily get to the joints to clean them e.g. BGAs.

>>1234850
Capacitance sensor below the disk.


Heeeh heeee I am a bad personnnnnnnn.

Kester 245 or 44?

>>1235592
245 is no-clean. 44 is RA, stronger than RMA. See
>>1235561

>>1235557
>you mean when the mains voltage is transmitted directly to what was supposed to be the LV DC lead attached to my project

Not necessarily mains, but any short. Literally lead to case to another lead; same as if you shorted random contacts on the board with a jumper wire.

An insulator under the board is a big deal when mains is involved, less so if it's isolated at the socket (or at least somewhere outside the case). It's still a good idea, though, especially given that it's 0.2¢ worth of material and minimal effort.

>>1235348

For constant speed you will need to measure the speed and use it in a feedback loop. You could perhaps consider using stepper motors, if your speed/torque requirements aren't too crazy and you can live with having discrete steps.

Exception is if the load on your output shaft is constant.

I need to have a uc without a real-time clock get samples from an ADC at a fixed sampling frequency. My current plan is to figure out the number of instruction cycles needed to grab each sample and then wait a certain number of cycles in between each samples to get the right period. Is there an easier way?

>>1235685
How about using the timer that's already there and probably already counting CPU cycles?

>>1235703
Thanks, looks like that's the move.

>>1235370
Assuming the motor has constant impedance and isn't drawing a significant fraction of all the current the 9V batteries can provide, doubling the voltage will quadruple the electrical power, which can be expected to increase motor speed or torque, depending on the load. The effect of the sagging voltage in series will still have the voltage higher than from two cells in parallel unless the motor has a really low impedance, and hence it will have higher current.

>>1235720
>Assuming the motor has constant impedance
which is hasn't, impedance depends on torque. Using 9V batteries for motors is inefficient. Use 6xAA instead.

>>1235738
Well assuming we have a significantly small motor, doubling the voltage by putting the batteries in series at the expense of half the current-carrying capability won't make a shred of difference, and I can't imagine that the impedance will change by a factor of two. So is this not a more reasonable assumption to make than assuming that the voltages of the 9Vs in series are sagging to below half their parallel voltage?

File: instruction manual.png (67KB, 391x432px) Image search: [Google]

67KB, 391x432px

>>1234666
Electronics noob here. I got a steering wheel media control setup, and the instructions aren't making sense to me (thanks china). I've heard people say they just soldered it all to a 3.5mm jack, but none of them go into detail what wire goes to what connection, any input? Pic related, the instruction guide with the wires.

>>1235746
>at the expense of half the current-carrying capability
Where did that come from? Batteries still supply the rated current when stacked.
>I can't imagine that the impedance will change by a factor of two
You can't imagine that putting two batteries in series will double the ESR of the pair?

>>1235795
Need pics of whole device.

File: 61gE0tFcWTL._SL1000_.jpg (79KB, 1000x1000px) Image search: [Google]

79KB, 1000x1000px

>>1235802
Here's the complete kit. The controller itself sends a signal to the box, so no wires to the actual controller. It's just a matter of what goes where when trying to attach it to the jack. I have more pictures of the connection between the wires and the box if needed, but I don't think that's relevant.

File: 1492719249569.jpg (146KB, 800x800px) Image search: [Google]

146KB, 800x800px

>>1235808
ACC is the accessory power, which should be available from your existing radio. GND is available anywhere on the frame. Control wires go to the "control interface" wires on your GPSnav device, they should be marked 1 and 2 respectively. Antenna line remains disconnected.
You should probably reference a wiring diagram of your car from a repair manual while installing. If you don't have a repair manual, get one or look online.

>>1235802
You misunderstand. For starters, if his motor has similar or lower impedance to the ESR of the batteries, then 9V batteries are a bad choice. In regards to the first point, I meant that two batteries in parallel can supply twice the current as two in series, and for the second point I was referring to the impedance of the motor. I've never heard ESR of a battery or other DC source referred to as impedance.

>>1235170
Hey, thanks.. i did the stuff you suggested (did not replace caps, didn't look bad/didn't have replacements anyway), and it worked out great. 20 year old amp now sounds like 10 year old amp again

Has anyone ever heard of someone taking an old cell phones camera module, and using it for a project.
It seems impossible from what Ive read.
I know its no simple feat yet Id be willing to really dev into this and learn, but every post I read just says its impossible.

>>1235849
mass produces camera modules are the most poorly documented thing you'll ever find.

>>1235854
Yeah but like, people have gotten linux to run on a ps4, I figured this would have at least been done once.

>>1235815
Thanks, appreciate it!

>>1235849
It would be a lot of effort, especially relative to the value of the part itself, and not necessarily reproducible.

>>1235847
>>1235861
No problem.

>>1235817
That's fair. /ohm/ takes all kinds, even those who say amperage. Terminological correctness is never guaranteed.

File: transformer.jpg (3MB, 1920x5580px) Image search: [Google]

3MB, 1920x5580px

Not an electronics project that involves any thinking or requires serious knowledge, but I took an old transformer that had four 6 V secondary conductors, and used some relays to switch between all of them being in parallel for 6 V, two pairs of them being in parallel and those pairs being in series for 12 V, and all of them being in series for 24 V. The voltage can also be adjusted in smaller steps through the primary conductor (~4.5 V, ~5 V, ~5.5 V, ~6 V, similar for 12 and 24 V). I mostly use it to power "cordless" drills whose batteries are fucked, and my hydrogen generator.
The diodes get almost unbearably hot when operated at 8 A, and get really hot when operated at 15 A (they stay below 100 °C though), even though I used thermal paste, and the things were rated for 30 A with a heat sink. Does anyone know what's the maximum temperature for these things? It's not written on them.

>>1235907
Never mind, googled the model number, max temp is apparently 150 °C and they are rated for 25 A, so I guess I'm pretty safe.

File: 6SA7.png (244KB, 873x855px) Image search: [Google]

244KB, 873x855px

Does anyone have a SPICE model for a 6SA7?

Just bought this as my first scope:

https://www.aliexpress.com/item/FREE-SHIPPING-Hantek-DSO5102P-Digital-Oscilloscope-100MHz-2-Channels-1GSa-s-7-TFT-LCD-Record-Length/32564105218.html

Apparently hackable to 200mhz. Hard to beat that price. Will I regret it?

>>1235917
>simulation program with integrated circuit emphasis
Good luck. Maybe you can pretend it's a JFET.

>>1235918
Yes. You would have been better off either buying a vintage analog/digital scope or forking over the extra $200 for a Rigol DS1054Z.

The late 80s and 90s era digital scopes from Tektronix and HP/Agilent are top tier (they were the $10,000 scopes of their time) a lot of them do over 100MHz bandwidth easily and have all the necessary features a beginner would require and often more. My HP54645D cost ~$200 (you can find them cheaper) and has an integrated 16-channel logic analyzer and the option that gives you derivatives, integrals, and FFT. It's 100MHz 200MSa/s. They're very robust too, not made of cheap Chinesium. I have a Tektronix 2430A that cost closer to $100 as well which is also a very capable unit in it's own right. Has a higher bandwidth but no functions like FFT and integration. Does have 2 external trigger inputs though vs the HP scope which has none in order to make room for the logic analyzer.

>>1235849
Has anybody tried to make an open-source hardware mobile phone, akin to the TS-100 soldering iron? If so, you might find a mass-produced tiny camera module made for them that has easily accessible instructions for hobby programmers.

>>1235907
That's a bunch of secondaries, where did you get that sort of high-power transformer from? Having a different fuse for each line is a good idea. All in all a pretty cool project, reminds me of Matthias's bench top power supply that uses what I think is a "tube tester" transformer with a bunch of taps.

>>1235929
I got it from a relative who has a bunch of junk lying around in his house, I would assume that it was made in the communist era, and was most likely a lead-acid battery charger. The secondaries were originally wired in series, so they didn't really give a shit about getting the most wattage at every voltage, which would make sense for a household battery charger.

>>1235922
>You would have been better off either buying a vintage analog/digital scope
To be fair, it has to be a rather nice vintage scope to beat that Hantek's cheapo scope. Just a couple threads ago some anon went that route (4ch 100MHz HP digital scope) and it wasn't exactly a great buy.

>>1235929
>Has anybody tried to make an open-source hardware mobile phone
Yeah, but what he originally asked (reverse engineering some common cell phone camera) has been done several times. Also, sometimes the full datasheet gets leaked. For example, you can get the full datasheet for Omnivision's OV07740-A32A ($6 from Digikey).

>>1235921
SPICE tube models are available and some of them are pretty good. They're mostly for amplifying tubes, but some mixer tube models exists. Probably not for his particular tube, though.

File: IMG_6088.jpg (151KB, 1106x902px) Image search: [Google]

151KB, 1106x902px

>>1235922
>Chinesium
Yeah I get this but reviews were good enough for me. I thought getting a vintage scope and might still get one if I spot a good deal. So far prices of working vintage scopes were relatively high. Perhaps due to school.

File: sales-meme-cowbell.jpg (44KB, 400x400px) Image search: [Google]

44KB, 400x400px

>>1235907
>The diodes get almost unbearably hot when operated at 8 A, and get really hot when operated at 15 A

From your pic I'd say, "Needs more heatsink."
Find a piece of aluminum the size of the board everything is mounted on.
Move everything to the aluminum plate.
(don't use the small, finned heatsink)
Use thermal paste for the rectifiers.

All-in-all, looks like a very effective battery eliminator.

>>1235950
I've looked it up since, and apparently it's normal for them to get hot. I don't really have a bigger heat sink anyway, and even if I did, it wouldn't fit inside the box then. The rectifier without a heat sink doesn't get hot btw, because it's only connected to the fan and the relays, which draw less than 1 A combined.

File: derating_tool.png (42KB, 340x163px) Image search: [Google]

42KB, 340x163px

>>1235908
>max temp is apparently 150 °C
Junction temp, not surface. Look at the derating curve.
Pic: What tool is this?

>>1235953
>I don't really have a bigger heat sink anyway, and even if I did, it wouldn't fit inside the box then
>>1235950
>Find a piece of aluminum the size of the board everything is mounted on.
>Move everything to the aluminum plate.
The size doesn't change, just the material.
(not a heatsink, just a flat aluminum plate)

If you're happy with it that's all that matters.

File: 31GMnJxHvnL._SX425_.jpg (7KB, 425x319px) Image search: [Google]

7KB, 425x319px

Hello,

It seems I got trouble with some GU5.3 / M16 connector. If I move a bit the bulb in the socket light comes in, then disapear after some times. The connector might be lose or something.. How could I tighten the connection between bulb & connector ?

Everything is fine with the powering, I got a small 220V AC => 12V DC, 20W power supply. Putting my multimeter in it, the needed voltage is there.
The bulb consumes 8W and works well while testing it on a bench.

>>1235949
You really should at some point. They are very worthwhile and it shouldn't be too much trouble to find a decent deal. Give an old scope a new home so it doesn't languish in a dump somewhere.

File: file.png (67KB, 559x423px) Image search: [Google]

67KB, 559x423px

>>1235958
>Pic: What tool is this?
It's a handheld torch for lighting the balloon. This model is kinda shit, the metal piece that pushes the piezoelectric igniter inside when the button is pressed is prone to breaking, because it's made of some very low quality metal.
>Derating curve
I haven't heard about those before, so I guess posting here was really useful. Pic related, mine is the GBPC25. So if I understand it correctly, the max current flowing through the rectifier shouldn't be more than 20 A when its temperature is 75 °C, for example, otherwise it will degrade more quickly, right? I guess I was still safe at 15 A, but I was pushing the limits a bit. I suppose my best bet would be getting the 35A one, because it's unlikely that I'll achieve better cooling efficiency, although there was no airflow when I did the temperature test, because the whole thing was outside of the box.

>>1235959
Oh I see.

>>1235966

there's no way to tighten those suckers without breaking them. what you can do instead is make the light-bulb pin fatter with solder or by jamming an extra wire in the hole.

(the solder trick is not likely to work coz they tend to use some metal to which solder does not adhere, my dear.)

>>1235973
>make the light-bulb pin fatter with solder
Fair point, thank you I will go for this.

Anyone know a good source for cheap stripboard online? Just the basic strip kind, I ain't no city slicker who needs all kinds of fancy pads and and things

>>1235947
The reason I asked about cell phone cameras is because the modules are of much higher quality and cheaper than stand alone ones.

File: pick and place.jpg (3MB, 5312x1706px) Image search: [Google]

3MB, 5312x1706px

i made a vacuum pick and place tool. i wanted to test the grip and it works so now i'm going to make a vacuum pump and solenoid enclosure for working it.

File: strip_dot_board.png (136KB, 190x282px) Image search: [Google]

136KB, 190x282px

>>1236065
www.google.com/search?q=stripboard
I mostly use smd-friendly dot board for prototyping.

>>1235967
Where are you going to find a vintage scope that beats out modern Rigols or Hanteks while still being cheaper?

>>1235918
You did fine.

>>1236153
>solenoid enclosure for working it.

I'd suggest just drilling a hole in the body, near where you grip it.

Cover hole with finger to grab. Uncover to let go.

yo, for high power LEDs (1W-20W), there is no need for current limiting right? a series resistor is unrealistic in this case right?

>>1236546
Why wouldn't you need one? They're still going to draw huge amounts of current if you don't have one.

>>1236547
well in high power applications resistors become unrealistic because of the amount of waste power generated on I^2*R. i was basically fishing for if you need like an impedance transformation or something, idk?

noob here, any suggestions for an adjustable power supply. something under 20$ from china would be best, since cash is not plentiful. I don't need much since i'm mostly playing around with arduino and similar.

>>1236550
So have a smaller resistor. Or build an active current limiting circuit. If you just hook an LED up to a voltage source it's going to let the magic smoke out.

File: Screenshot_5.png (137KB, 1327x751px) Image search: [Google]

137KB, 1327x751px

Guys, where do I Hook up the LVP (Low Voltage Programming) in a PicKit3? What does it do?

Pic Related, trying to use it in a PIC16LF876

File: 71NfVbsyrjL._SY355_.jpg (16KB, 355x355px) Image search: [Google]

16KB, 355x355px

>>1236564
>an adjustable power supply. something under 20$

not going to get much for that.

adjustable power supplies are a luxury that you seldom really need. for an arduino you can get one or more DC power adapters and do lots of things. Seldom do you really need to turn a knob and change your power supply from 10 volts to 11 volts, or similar.

also, they make DC power adapters with a switch that changes from about 3 volts up to 15 or so, and these are inexpensive as well. regulated power supplies cost more, but again, you can do lots of things with unregulated adapters.

>>1236576
My guess is you don't do anything with it unless you have it set up for low voltage programming, which sounds like something you wouldn't really need to do. I made a USBPICProg, and it works fine with only Vpp, Vdd, Vss, PGD and PGC.

>>1236576
just leave it floating

File: 1493646234361.jpg (72KB, 1024x768px) Image search: [Google]

72KB, 1024x768px

>>1236577
thanks for the suggestion, already found one that's be perfect for me

>>1236550
>in high power applications resistors become unrealistic
Correct. You add a switching current regulator, of course.

>>1236654
oh ok, duh. thanks senpai

File: 1476384895653.png (107KB, 701x904px) Image search: [Google]

107KB, 701x904px

>>1236678
Pic related, for example.

>>1236537
i'm avoiding that because it requires hand movement so i'll be using a foot pedal switch. i could use a pneumatic pedal but that's just more hoses going everywhere.

>>1236583
>>1236578

Many thanks.

What does it do though? when would I want to use it?

>>1236733
LVP is just a different protocol for programming the chip that is enabled by a logic input pin (PGM) instead of a high-voltage input pin. Use it when you can't get 13V Vpp to the /MCLR pin because of e.g. other circuitry that won't tolerate it. See the separate "programming manual" for your chip for more details.

I'm completely new to electronics on a component level, all I ever did was use/modify prebuilt stuff.
>The (stupid) question:
I want to essentially manually control the Windows sound mixer with a set of physical sliders. Here's what I thought, I want to stay as cheap as possible without it being absolutely shit.
I thought a ATMEGA328p-pu could read the resistance of slider potentiometers. Alone that should be fine. How would one go about transfering the data from the Atmega to the PC and controlling the sliders?
I can write a little Python and JavaScript, that's about it.

>>1236782
Or use an analogue summing amplifier? Why not just have the thing plug into your 3.5mm socket(s) and be simple like that?

>>1236792
He's talking about the mixer that controls how the sounds are summed in windows before they're output.

>>1236782
I don't know nothing about no Windows audio APIs, other than that they're there. If I had to do this, I'd start with a program like MVcontrol http://dev.strobotone.de/ that controls volume as long as I send it messages somehow. This one happens to accept MIDI messages.
To generate the MIDI messages, you probably would want a microcontroller that has real USB support built in such as the ATMEGA16U2 (e.g. Teensy board), because USB is complicated and you can use a framework to increase your chances of success (I like LUFA). There is an example for almost any USB standard device class, including among others a MIDI controller example. You still have to learn a bit of C to pull it all together on the device side, but if you've programmed Arduino before this shouldn't be much.

>>1236796
Not sure if windows could do that, but some form of linux probably can. But what would you do when you move the slider on the computer and the IRL slider is stationary?

>>1236867
Usually, the next touch of the physical slider causes a jump to the physical slider's location. What else would you do?

>>1236869
I guess so, but it would be good if you could disable the software side when it detects the gadget being plugged in.

https://controllino.biz/

>Arduino-based PLC
What a time to be alive!
>Starts at $125
Well nevermind I guess

File: 3 sensors 2 inputs.png (9KB, 564x433px) Image search: [Google]

9KB, 564x433px

So I want to connect two switches and a photo transistor (SFH 309) to two inputs. I made myself a truth table to warp my head around it and came up with a circuit.
Do I need the diode I drew in red to make sure the marked row in the table is correct?

>>1236988

you do not need the diode.

>>1236973
saved
Im Impressed of what it offers, the biggest turn of is it being Arduino based and not having
"real" plc programming software.
Hardware wise it is god tier in comparison to other small crap controllers like Siemens Logo and such

>>1236992
why? Will the phototransistor act as one?

>>1236973

$125 for a PLC, if it's good, is nothing

>>1236994

yes. that's why it is drawn the way it is.

>>1236988
A reverse-biased transistor operates like a Z-diode with a poorly definded avalanche voltage. With 9V at I2 you may find a volt or three at I1. I would include the diode if that matters.

>>1237009
It's not only the avalanche effect. Transistors do work in reverse, just much less effective. Simple experiment: set multimeter to ohm range, connect phototransistor, note resistance at a certain illumination. Reverse polarity and compare.

>>1237020
Diode Y/N?

Hi /ohm/

I'm trying to build an arduino controlled battery welder but I've got a problem

I can't for the life of me find a proper solenoid relay so I'm forced to use automotive ones.

Question is, since I need to push about 100amps, tops, and I have a handful of 40A capable relays, can I just set like 3 or 4 relays in parallel to be about to hit the amperage mark without melting the relays contacts shut?

>>1237036
Y

>>1237068
where are you getting the current from?
cant you just switch the input Mains on the primary transformer side like an ordinary welder does?

>>1237068
If there is no other way then do it. Be aware that no two relays close or open their contact at the exact same time. The more you have in parallel, the better. Will require substantial coil current, don't forget the clamp diode(s) for inductive transient suppression. They must be able to handle the same current as the relay coils. Can you not switch the mains to the transformer instead? Snubber network and all..

>>1237078

I am FAR from qualified enough to dick around with mains power, I'm using a fat 12v car battery for a power source.

>>1237106
>I'm using a fat 12v car battery for a power source.

that shit will burn you more than mains, which has a breaker.

4chan. you gotta love it.

>>1237110

All I'm doing is using an arduino to trigger a relay for a chosen amount of time, the 12v being relay'd so to speak.

>>1237120
Actually thinking about it now, I won't even pull 100amps, and I forgot that a motorcycle battery will suffice, no way they deliver 100A

>>1237120
>the 12v being relay'd so to speak.

all I'm saying is that a car battery can deliver hundreds of amps and won't complain until it is dead.

carry on, anon, you seem to know what you are doing.

File: 322895-might_guy.jpg (11KB, 320x240px) Image search: [Google]

11KB, 320x240px

>>1237122

Thanks for your support!

>>1237121
A motorcycle battery will absolutely deliver at least 100A.
A relay will not pull anything close to that, but a short circuit will.
Use appropriate fuse and take care.
Have something insulated on standby to beak the circuit when testing. Not for electricity, but because dead short red hot conductor can burn the skin off your hand before your brain can register the pain. Find somewhere well ventilated if you are playing with lead acid batteries.

>>1234850
Get a few Peizo disks, they're like 10 for a buck or something. Rig up a simple signal amplifier to it (since the voltage coming out of it is small), you can make it either using an OP-AMP or just a couple of BJTs and use an OP-AMP as a comparator to set your trigger point.

I built something very similar like 4 years ago for a class project, I might have the schematics lying around somewhere if you can't figure it out.

>>1235151
Depends, capacitive coupling only happens with AC signals, if the node has active AC filtering (like a DC voltmeter) the floating voltage change isn't capacitive coupling.

>>1237068
I wouldn't. The relays won't be guaranteed to close at the same time, and one of them WILL be passing the full 100A for several microseconds at least. How about a MOSFET or three?

>>1237121
>no way they deliver 100A
>what are cold cranking amps

>>1237068
>FQP60n06
>60 volts VDS
>60 amps continuous (with sufficient heatsinking)
>Turns on in under a microsecond so no problems with timing uncertainty like in relays.
>2-3 volts Gate threshold voltage, less than 20 uAmp gate current. Can be driven directly from an Arduino without needing any sort of intermediate transistor, which you'd need for a relay.

Get 3 or 4 of them (they're like 2 for a buck or something) and a big ass heatsink and stick em in parallel and you're good for 100 amps. Make sure you try and get some from the same batch since it can still be problematic to have ON-resistance mismatching which might cause one of the mosfets to bear more load and blow.

>>1237163
Why not irf3205, very common and can handle 110 amps.

>>1237232
Personal preference. I just really like the 60n06 and use em for whenever I need a moderately powerful Mosfet. That and the fact I bought a hundred of them for some reason. Point is, it's much better to use Mosfets than relays

>>1237232
The 3205 isn't logic level, for one thing. IPD031N06L3 is, at least for 5V CMOS, and it's fairly cheap.

>>1237110
12VDC is dangerous in a short circuit due to instantly producing a lot of heat, but mains voltage is dangerous without warning in any kind of circuit. If you touch the wrong piece of metal in a mains circuit then you're not going to have a nice time, but you can feel free to put both hands across a car battery without any risk. But that's more because DC is less dangerous as opposed to the voltage itself.

>>1237145
Not him, but how linear are those piezo discs? Don't they generate voltage proportional to changes in deformation as opposed to deformation itself?

>>1237296
>DC is less dangerous

Yes, Mr. Edison...

>>1237296
Even mains isn't as dangerous as it's often made out to be, at least not the 120V in America. It depends a lot on how you come into contact with it. If you only brush against it with a finger or hand and any connection your body has to ground is relatively high impedance you'll almost certainly be fine. You will definitely feel it but you won't be any worse for wear. If on the other hand you were to grab say a live bus bar with the right hand and the neutral bus bar with the left with a good grip with good contact you'd be fucked. You wouldn't be able to let go. Almost no one dies from mains electrocution, certainly no one just poking around in an electronic device. Most people die when they are working on a house or something, hit a live connection, and fall off a ladder or from some high place. Those that are actually electrocuted off mains are either exceptionally stupid or exceptionally reckless.

>>1237298
12VDC won't kill anyone unless you take the electrodes and jam them directly into the heart. Even higher voltages like 100-200VDC can be touched and while painful aren't lethal in most circumstances. Once you get up into the kilovolt and megavolt range DC is just as dangerous as AC if not more. AC generally can kill you at lower voltages though. It has to do with capactivie and inductive reactance of the body lowering the overall impedance (Z=Xl+Xc+R). With DC only resistances, not reactances factor into the impedance of the human body (Z=R).

>>1237301
>It has to do with capactivie and inductive reactance of the body lowering the overall impedance (Z=Xl+Xc+R)

Actually I want to rephrase that because that's not really correct. The reactances of the body have a tendency to provide lower impedance paths through the body than would be available at DC.

Inductors look like a short at low frequency and capacitors look like a short at high frequency. One current path through your body might have a total resistance of 100kohm but an LC equivalent current path parallel to it might only have an impedance of 10kohm and thus will shunt more the current through that path inflicting a whole lot more damage to your body. With DC the only current path is the 100kohm path reducing the amount of current passing through your body.

>>1237305
this is correct and especially relevant at high frequencies, but for mains frequency (<400Hz) the body's capacitive reactance is large relative to its resistance so it doesn't really matter. it does mean touching a high frequency switching node is extremely bad. i believe the real distinction between otherwise equal ac and dc is the difference in contractive behavior you get with muscles at dc versus low frequencies.

>>1237309
Also it's the capacitive coupling between your body and ground. You touch a mains wire in bare feet you get a shock, even though there's no physical connection to ground. That's probably not going to happen with the same voltage at DC, though I'm just guessing.

File: tubes.jpg (196KB, 1023x583px) Image search: [Google]

196KB, 1023x583px

>>1234666
Cleaning pots.
I got an old oscilloscope to try and rehab.
After I clean it out and start looking at shit, I think I already figured the intensity pot is fucked. It looks fine, super bright if I get the intensity knob at like 95-100%. And there's one spot around 25% that works, otherwise there's no picture. That sounds like a dirty pot to me.

That said, I have never had contact cleaner do shit for me. It there anything that's actually good at this? How do I get it in the pot?

>>1237322
I'd just replace it, not hard to do and it won't look any different, but it's up to you whether you actually want to do that.

>>1237322
If sprays do nothing, you could open the pot and clean it manually. Also, if it works only at one point + at the end stops, there's a chance there's something wrong with the wiper.
Failing that, there's >>1237329's option. IIRC some old analog scopes require high voltage pots for that purpose, though.

If I put music into a 8 bit ADC which feeds into an 8 bit DAC, what would it sound like, assuming I have a MHz clock?

>>1237418

it would sound like a Nintendo, of course.

however speech recording is actually quite respectable at 8khz; it just sounds phone-like muddy.

it may be possible to simulate this with a program like Goldwave, coz it lets you record PCM 8-bit stereo, among others.

>>1237418
rencode the soundtrack to 8-bit mono and whatever samplerate your adc support. For example with Sox:
sox input.wav -r <samplerate> -b 8 -c 1 out.wav
That will be the best result in ideal conditions, depending on the adc the result may be much worse.

>>1237418
It would sound like www.google.com/search?q=8+bit+music

>>1237425
>>1237431
Would it though? 8 bit music seems to only have 2^8 (just a guess) different notes, hence the music genre's simplistic sounding nature, but that's a midi thing. The digital conversion (assuming a perfect conversion) would just be limiting the amplitude of the voltage to certain intervals, which would distort waveforms. If your sample frequency is high enough, you'd be able to hear any note you wanted with a degree of distortion, provided it wasn't too quiet, right? The quieter a note was the more distorted the waveform would be until it was either a square wave or not audible at all, but how bad would audio sound at high amplitude?

File: Screenshot_20170902-143344.png (3MB, 1080x1920px) Image search: [Google]

3MB, 1080x1920px

hi guys, can someone help me with these simbols, what do they mean?

File: 8vs16bit.gif (135KB, 1469x828px) Image search: [Google]

135KB, 1469x828px

>>1237458

if you mean the circles and arrows: sept 2011 (09 - 11)

>>1237446

the steps are larger at the top (8 bit) compared with 16 bit at the bottom. sound is not that bad, but the distortion gets annoying after a while.

File: $_57.jpg (128KB, 809x669px) Image search: [Google]

128KB, 809x669px

Cheap FPGAs?

I need something that's easy to develop in and has the best chances of having a good development life.

Any ideas or websites that might have one?

>>1237446
"8-b music" just refers to the techniques typically used in 8b computers. Usually they had a few (3 for example) programmable generators for producing the notes and a noise generator. Sampled music wasn't usually used, due to high memory and processor power consumption.
You could answer your own question by using any random sound editor by saving your file as 8b and then playing it back. Often the most obvious change is the significant increase in noise level.

>>1237473
For altera FPGAs you could buy a terasic board:
terasic.com.tw
You can get a cyclone IV or V for ~60$ or ~100$ respectively (they also offer a student discount).

The equivalent choice for Xilinx would be digilent:
store.digilentinc.com/fpga-programmable-logic/
you should be able to get an arty or spartan7 for the same price range.

You can find most of those models also on websites like digikey and farnell.

Most boards will have an integrated usb jtag and full compatibility with the respective official development toolchain (quartus for altera and vivado for xilinx), so you just plug an usb cable in your pc and program/debug it with the official tools.

>>1237469
> if you mean the circles and arrows: sept 2011 (09 - 11)
yes, thank you

in general, what's the difference between cheap fpgas and cplds for regular glue logic or high speed data applications?

>>1237583
The most visible difference from the design point of view is usually the very low number of flip-flops in CPLDs, often just 1 or 2 per IO pin.
CPLDs tend to have more predictable timings and support much wider functions (inputs per output) without slowing down.
The "program" is more or less permanently programmed to CPLDs, meaning that they start faster. Your supar sikrit donutsteel is also better protected.

IMO CPLDs are nice for relatively small, straightforward designs with low number of FFs. If your design does not fit to that category, FPGA might be a better option.

File: edge detector.png (13KB, 1121x822px) Image search: [Google]

13KB, 1121x822px

I want to get triggers from a pulse, and I want them to be relatively slow/long, since this is going to simulate a button press and needs to get around the debounce time. I also want triggers on both the positive and negative going edges, so I came up with this. The idea is that input 2 goes high while input 1 is low, meaning the output goes high, charging the capacitor, which sends input 1 high, switching the output off, and the capacitor discharges. Will it work? I don't know, and since I'm using soic chips I'd need to build breakouts to breadboard them so i'd appreciate some insights.

If I want to measure resistance with a Arduino nano, I have to do so via measuring current from the analog input right?
So I measure the resistance of a potentiometer by measuring current and using Ohm's law: 5V=R*current => 5V/current = R. If I need at least 100 values (percent) and I know that arduino can measure 1024 values, does the resistance matter? Can i use a 1MegaOhm potentiometer instead of 50 or 10k ohms? Or does that destroy the resultion/not work?
I'm really new to all of this. Thanks in advance.

File: Fork20-18-18.png (182KB, 516x618px) Image search: [Google]

182KB, 516x618px

>>1237681

AVRs measure voltages at the ADC pins, scaled to 5 volts typically. Your arduino has an AVR.

You seem to know about Ohm's law.

Think about Ohm's law. a lot.

and why the ever loving fuck would anyone want to measure resistance with an arduino.

seriously person, you're smarter than that. move on to blinking LEDs and cool stuff.

>>1237418
A file decimated to 7 bits, approximately.

>>1237666
Probably not. A NAND gate is not a comparator.

>>1237681
Most microcontroller ADCs sample voltage, not current. Impose a known current through a resistor, such as the end contacts of a pot, and read the voltage from the resistor you wish to sample, say, on the wiper.

>>1237700
>can't into front panel controls
>thinks blinking LEDs are cool
gb2r

>>1237702
>Most microcontroller ADCs sample voltage, not current. Impose a known current through a resistor, such as the end contacts of a pot, and read the voltage from the resistor you wish to sample, say, on the wiper.
Thanks a lot for your help! Then this really wasn't an issue to begin with and I just didnt understand it completely.

>>1237681
y tho

>>1237702
don't suppose you have any other ideas to get rising/falling edge pulses? It looks like the canonical method is chaining up gates to create a delay but i'm going to be using 10+ instances of the circuit and would prefer not to use a whole chip on each one. tempted to just give up on doing it with logic and use an atmel 2313

File: edge detector.gif (25KB, 1296x1325px) Image search: [Google]

25KB, 1296x1325px

>>1237666
>I want to get triggers from a pulse

dont make up your your own terms. what you want is an edge detector, or ''short pulses from both rising and falling edges of a pulse''. for that you use a differentiator, preferably but not necessarily using Schmidt trigger inputs.

attached pic has 2 examples. the top design is good and solid, a classic: you could use inverters like the 74c14 instead of NAND gates. if you want simpler, you can try the bottom design, which i havent, and wont, coz it's not as solid.

>>1237736
thanks for this. I'll try building that top circuit with inverters and fiddling with capacitor values

>>1234806
Your forgot the hands smoking too

File: 20170902_202459.jpg (3MB, 4032x3024px) Image search: [Google]

3MB, 4032x3024px

Can anyone give me a hand with this clusterfuck of an ethernet bus? The cable I care about is connected to the pins on the top right. I connected brown and brown and white by the brown post, green and green white by the green post etc. I've tried plugging my laptop into the corresponding wall port with the cables there arranged in 568a and b configurations, and had no luck with either. The thing that's really throwing me off is that I can't find any consistent pattern to how the other cables are wired to the bus.

>>1237810
Picture rotated.
>top right
Leftmost, top side.

>>1237700
>scaled to 5 volts typically.
check the datasheet, what is typical is it can't be outside the voltage you are operating the chip at, there is device dependant sometimes an input that allows you to set full scale reference.

>>1237702
>Impose a known current through a resistor,
how do you propose to impose a known current through an unknown resistor without some wicked cool current source?
you need some kind of potential divider which is actually what a potentiometer is or a wheatstone bridge type dealie but in both cases you put a known voltage across it and read off a voltage too.

>>1237681
to actually answer the fucking question anon, the resistance/impedance of your circuit is absolutely important, the microcontroller will have on board a capacitor that is charged up from the input pin and voltage you feed. basically if you put too big a resistance on the input pin it all can go tits up. you know what you should do? read the datasheet, it will all be in there.
for completeness a 1MR pot will in magic theory cuckoo land give the same results as a 10kR pot if you disregard that 1M will be too large for the ADC to handle (again, read the datasheet). check out the calculations for potential dividers, its all about ratios. larger resistances will however be more robust, any impedance you add to the output of a voltage divider will change the outcome because the resistances have changed, say you are smart and add an opamp as a buffer, ideally the opamp has infinite input impedance, voila right? hasn't altered the reisistance? except in the real world its not perfect it has a huge resistance but not infinite. the larger the resistors in your divider the less it is affected by the imperfect impedance? is that right? fuck im so wasted right now. anyway it will certianly use less current and say drain a battery more slowly. less quickly? both.

>>1237810
you know this is a patch panel right? (i hope) so what is connected where? you are connecting your laptop to the visible wires? through a socket or the wires are actually a cut cable?

what should happen is:
1) ethernet cable from switch into back of patch panel in your pic, make sure managed switch is configured correctly.
2) solid core cable punchdown into back of patch panel. i think cat5e has 2 pairs of 2 pairs of wires, two of the colours are twisted at one rate, two are twisted at a different rate. it kind of should matter but sometimes idiots get lucky. match the colours and remember what scheme it uses, A or B. keep as much twisted as possible.
3) at the wall port wire it up in the same scheme as the panel
4) patch cable to device.

so if i were you i think check that the patch panel socket you are interested is connected to anything? i'm not saying you are retarded but these things happen lol.
if its not a mouded patch cable and instead made in house check the pinout at both ends, someone might be fucking with you.
make sure patch panel and wall socket follow the same scheme as each other. the wall socket should come off to reveal the same punchdown stuff.
those brown pairs doing there own things might be a huge concern, where the fuck are they going? are you sure this is a computer network? not some weird telephone hybrid or something...

>>1237480
Spotted the intelfag. When will my stratix 10 actually hit production??

Get a Zybo or zedboard

>>1237825
Yeah, the cable's running to the patch panel go to individual wall ports. Every wall port in the building works fine except the one in the room I'm working on. This infrastructure hasn't been serviced in god knows how long and the brown pairs for that cable weren't connected to anything.

The sockets on the other side are all filled and going to the switch, which also seems to be working fine.

The brown pairs are going to cables, they're just outside the zip ties for whatever reason.

I guess my issue is that I don't know how A and B scheme pinouts work on the panel. There's nothing to identify the pins except those colored markings, and I don't know if the markings were made with an A or B scheme in mind, or something else entirely.

I'm also thrown off by the wiring of the other ports. All the ones on the bottom right side have only three pairs to a port, and the wiring by color is different from how the blue cable is connected.

>>1237836
it looks completely fucked desu, whoever wired that is a savant or colour blind.
if you mix up A and B it really just changes a straight through connection to a crossover, pretty much everything in the last 20? years is autonegotiating built in, it doesn't care if its crossover or straight through. the only difference between a and b is that ornage and green are swapped over, which is visible on some of these ports.
for 10/100 you don't actually use brown/blue so sometimes it can be wired up as a second 10/100 connection or for a phone line which can go some way to explaining odd things sometimes, might help explain why some of your brown/blue pairs are swapped as someone tries to swap A to B on a second pair down the same cable?
don't explain why in my opinion anyway the solid/wstripes are all fucked up.

it all depends on state of the wall socket, if you match the colours up it should work.

of course you might have wasted your time if it's actually broken because a rat chewed through it or someone hung a picture and split it with a nail etc etc.

>>1237841
Alright, thanks. I've got an ethernet continuity tester coming on Monday. Hopefully that'll help.

https://m.aliexpress.com/s/item/32807257731.html?trace=wwwdetail2mobilesitedetail&productId=32807257731&productSubject=1450-Pcs-Connector-Kit-2-54-mm-PCB-Pin-Headers-Box-Packaging-For-Arduino-Dupont-Electric&spm=a2g0s.9042311.0.0.yEZtKp

Grabbed these and a crumper for an upcoming project, what size wire do these take? Solid or stranded better?

>>1237822
>how do you propose to impose a known current through an unknown resistor
Some wicked cool current source. Or, in the case of a potentiometer, a known resistance.

>>1237850
22AWG may not fit if it's thickly insulated. Stranded wire can be crimped more strongly and is easier to work with. If you have the crimp height set wrong, you may need to give the insulation end of the crimp a gentle squeeze with pliers to dress them to fit into the shell. Use a small pick to lift the retaining latch if you need to remove pins from the shell.

Does a lithium ion protection circuit limit the current simply by measuring the current, or does it measure the change in voltage with respect to time, or some other method? What about lithium ion charging circuits?

I was thinking that the dV/dt of a cell would give a good representation of the current going in/out of it scaled by some factor proportional to the cell's capacity, which would be helpful in making a protection or charging circuit that works for any capacity cell/array of parallel cells.

>>1237958
>simply by measuring the current
This. Low-value resistors are comparatively cheap, and overcurrent in charging or discharging absolutely must not be allowed to happen, so second-order effects aren't of much use.

File: orangepi_usb.jpg (376KB, 962x856px) Image search: [Google]

376KB, 962x856px

I'm trying to get an extra USB on my orange pi one, so I've connected data lines to pads #3 and #4, and then used 5v and ground from the GPIO pins and connected these all to a new female USB port

but I can't seem to get the OS to recognize that something has been plugged in to the new USB port
I tried swapping the data lines around in case I got it backwards, but still nothing
am I missing something?

>>1237985
Armbian? You'll probably need to load a device tree overlay to enable the port in the OS.

>>1237970
So there's no generic charge circuit to use for an arbitrary capacity cell. Is there a TP4056 style circuit with a trimpot on it to adjust the current?

File: 1479650556944.png (76KB, 623x657px) Image search: [Google]

76KB, 623x657px

>>1237993
Sure, the TP4056 will charge just about anything Li-ion. Just change the programming resistor and make sure you don't charge too quickly. There may not be any boards so configured, but it wouldn't be hard to build one.

>>1237993
TP4056 has a resistor for adjusting current. Replace it with a trimmer?
Plenty of other similar charging ICs exist.

>>1237996
So a pot on those could work? Neat.

>>1237989
Yeah, armbian
I've tried looking into this
but I don't seem to have any of the relevant files
there's no /boot/armbianEnv.txt and there's no /boot/dtb/ directory
maybe the OS version I have is too old?

but anyway I don't think it should be required because the pads are right on the board and not part of the GPIO pins
but I could be wrong about that

>>1238031
>I don't think it should be required because the pads are right on the board
The Linux-ARM kernel maintainers disagree. If there is not a user-accessible socket on the baseline board, policy dictates leaving it disabled by default.
>no armbianEnv.txt
That's been there for quite a while. You probably should upgrade.
>no /boot/dtb
You might be running the default kernel instead of mainline. In that case, you would have to run bin2fex on /boot/script.bin , edit the appropriate ini section, then fex2bin it back into the script.bin and reboot.
>tfw doing exactly that right now to get some shit working
>iknowthatfeel.jpg

>>1237831
well, yeah I have more "experience" with altera but I also recommended one of the most popular xilinx partners because I didn't want to be a shill for intel.

>>1237917
Ordered some 24 awg stranded thanks anon

>>1237917
>Or, in the case of a potentiometer, a known resistance.
then you add your unknown resistor, now what is the current? known or unknown?

>>1238100
Dunno what he's trying to say, but if you put your known resistor and the unknown resistor in series, feed them from a known voltage and measure the mid-point voltage, you can calculate both the unknown resistance and the current through it.

>>1238102
the previous poster proposed feeding a known current through an unknown resistance and measuring the voltage drop across said resistor.

my concern was creating a known current was needlessly complicated compared to using a known voltage.

anon responded that a known current could be created by biasing a known resistor with a known voltage.

my point is that then adding an resistance, known or unknown would alter the total resistance and therefore current through your sample.

I have an old printer/scanner/fax machine that's busted. No chance of repair.
What stuff can I salvage from it? afaik the only thing worth taking are the steppers.

How much would you pay for a tek 475a with a dm44? I have found one on Craigslist that was pictured in a lot with some bk scopes, so I messaged the guy asking if he would sell just that scope, and he replied saying that it has become his personal scope but he'd be willing to sell it for the right price.

>>1238119
Power supply, scanner lamp, heater(if it was a laser printer), steppers and regular motors.

>>1238137
and stainless rods - can never have too many

File: pic.jpg (1MB, 1920x1080px) Image search: [Google]

1MB, 1920x1080px

i made an arduino-style microcontroller platform with an efm8sb2. it's my first successful project using 0201s and qfns. i have yet to program it.

tombstoning was a big problem. my first board, half my 0201 caps tombstoned. this time only one of the led resistors did but i managed to push it more or less back into place. i also burned the tiny button some. hot air straight up melted it the first try, and my soldering iron tip is like the fucking finger of god compared to these parts.

File: Still002.jpg (2MB, 1920x1440px) Image search: [Google]

2MB, 1920x1440px

>>1238141
close up before soldering. glare is kind of bad because of the flux.

File: ohm_meter.png (5KB, 250x250px) Image search: [Google]

5KB, 250x250px

>>1237681
You have a known voltage Uk, a known resistor Rk and an unknown resistor Rx. Connect Rk between Uk and ADC and Rx between ADC and GND. Now you can calculate Rx. The magic formula is Uadc = Uk*Rx/(Rk+Rx). Solve this equation for Rx and let your machine do the calculation.
>does the resistance matter?
Yes, that's why ohm meters have ranges. If you want a wide Rx range you need to switch Rk to maintain accuracy.

File: 1474091393509.jpg (29KB, 640x424px) Image search: [Google]

29KB, 640x424px

>>1238100
Potentiometer CCW-CW: known resistor
Potentiometer CCW-wiper: unknown resistor
I understand the equations are quite neat for that case. :^)

>>1238142
>0201
You absolute madman.
Bretty gud. The LEDs look a bit overbaked, tho. Try a hot plate underneath the populated board to get the board and components on it to 900C or so, and/or hold wide-angle lower-temperature hot air some distance away to get everything evenly warmed and drive off some water. When you bring the temperature up for the critical phase, you won't need to risk overheating small objects, or parts of large ones, to get large objects and the board itself reflowing.
Also if you don't do the hot plate, maybe place some washers/board scraps/bottle caps underneath the board at the corners to keep from having to warm the surface too.

>>1238119
In a laser printer I found a solenoid and the laser scanner potentially useful, in addition to the one stepper motor that ran the whole thing. I might, that is, would like to but probably won't replace the laser diode with a near-UV device and make a scanning UV printer for pc boards.

File: lsusb.png (19KB, 696x175px) Image search: [Google]

19KB, 696x175px

>>1238032
thanks for the tips
I learned some things

File: oidsajfosidjfoisdjf.jpg (18KB, 228x221px) Image search: [Google]

18KB, 228x221px

>2017
>not using a forth microcontroller
>freaking cavemen I tell ya

>>1238306
>Potentiometer CCW-CW: known resistor
yeah that is for sure a known current and in no way a known voltage applied....
what a jerk you are.

I'm building an arduino based gaming peripheral, and want to add LEDs which are supposed to lay in some frosted acryllic. If I want 4 LEDs to bei powered to light the 9*9cm square on the sides, what do I need to look out for in terms of brightness? Can they run on arduino only (40mA afaik)? How much do I need to go down with expectations with 40mA?

I have an LCD TV that's shit the bed a little; its control board is gone somewhere and for the life of me I can't find a shit-stain. A new board is elusive and at least $50 since it's an older model, and I was wondering if I can get a simpler control board that just accepts HDMI in so I can use it as a computer monitor instead, just in general? I know these boards exist for laptop screens, but what should I be looking for? Will the LCD inside the TV itself have a model number that the driver/control board will be specific to, or will the board be specific to the TV model or something else? Or will a general purpose board work?

A car alternator is basically a synchronous generator, right? Like, DC field in the rotor inducing a 3-phase alternating voltage in the stator?

>>1238420
Control board to panel will use some kind of LVDS that is dependant on the panel and controller. Like all cars run on fuel but you can't put gas in a diesel. If you can figure out exactly what the protocol is then you are golden. Your best bet which is a fucking huge long shot is finding a part number on the panel then getting a datasheet for it. You won't find a part on the control chip. Unless you might, but you won't.

>>1238312
Cheers m8

>>1238416
The best thing for you to do would be to find the highest-efficiency LEDs you can and do some experiments.
>40mA
OTOH, you might want to double-check that. You should be able to draw up to 500mA from a USB port in total, which is enough for lots of LEDs. Remember, you can always add transistors to switch more current.

>>1238420
>its control board is gone somewhere and for the life of me I can't find a shit-stain
I hate it when that happens.
The LCD probably has some sort of a part number. You can web search every number on the thing and see if you can find some data on it at panelook.com or something. If you're lucky they'll indicate that it's standard MIPI DSI and you can just plug it into any random LCD controller board on aliexpress.

>>1238412
It's a natural talent.

How can I approximate the starting cap for a 220V induction motor? I got a really old one and its starting cap is fucked, the writing on it is impossible to read. The motor is 1.2A.

>>1238460
>starting cap for a 220V induction motor
Are you certain it's a starting capacitor?
Many AC motors use run capacitors.
What is the application?

>>1238442
thanks for the answer, I meant 40mA for each pin, I got 12 unused digital pins (only using analog pins so far).
Thanks, I'll look around.

>>1238509
It's a multipurpose motor, probably around 30-40 years old, gonna use it in a bench grinder. I don't know anything in the motor other than what's on the motor's plate, which doesn't say anything about the cap.

Well I thought it was just a meme, but here's an MIT graduate unironically using the word "ohmage": http://www.cs.cmu.edu/~chuck/infopg/roboinfo.html

I suppose I can take solace in the fact that he is "just" a computer scientist?

>>1238608
Yeah don't listen to CS people when they talk about hardware. cf. "impedance match" when referring to database/API shit

File: 1486388999179.gif (160KB, 749x578px) Image search: [Google]

160KB, 749x578px

>>1234666
>>What books are there?
>Beginner
Any of these with some projects to go doing while you learn? I found this kind of way is the best for me to learn stuff.

>>1238643
The first one is filled with several projects.

>>1238643
go with Make: Electronics (2nd edition)
it's experiment first followed by explanation
starts from absolute beginner

>>1238559
I asked my electron plumber. If it doesn't start up well, C is too small. If hum while running, C is too large. Try 70 µF per kW.

>>1238716
Thanks, will do

>>1238716
ideally the reactance of the capacitor should be equal to the resistance of the stator, right? Or am I talking nonsense?

H-hi newfag here, i want to create a hand cranck generator, if i put 2 motors in series can i get twice voltage?

New retard here. Where do you guys get your PCBs? Do you pay the online service companies to print and ship them to you, have your own bought/diy pcb printer or something else? I'm thinking of fashioning a pcb printer out of some junk following a guide online. Is it a good idea?

>>1238774
yes, if +-+- zero if +--+

File: tesla64.jpg (12KB, 310x354px) Image search: [Google]

12KB, 310x354px

>>1238764
>the reactance of the capacitor should be equal to the resistance of the stator, right? Or am I talking nonsense?

if tesla were here he would put a bullet through your head first and then his

>>1238781
You'll definitely learn something if you want to undertake a project. Have realistic expectations about the quality of the results though.

>>1238781
I think etching is the simplest and cheapest method, for more complicated designs or several pieces, buying them from china.

File: Screen Shot 2017-09-05 at 11.06.03 AM.png (54KB, 818x930px) Image search: [Google]

54KB, 818x930px

Out with sketchup to design PCBs, in with GIMP!

File: pulsegen2_dia_pcb_s.png (21KB, 860x430px) Image search: [Google]

21KB, 860x430px

>>1238844
Nice. Population so far 1x DIP14, 2x R and 2x C?
I like dot board for prototyping.

>>1238619
>impedance match
The phenomenon is similar enough, even if it is qualitative rather than quantitative.
t.quite comfortable in both CS and EE worlds

>>1238781
Usually, I use a pcb pool like OSH Park. For more than 4 layers I'd try EasyEDA or one of the other whole panel companies.
If a simple, single-layer design I might use toner transfer, but I haven't really got that process all worked out yet. I hear you can use an LCD screen with a UV light source to pattern photosensitized pc blanks, which tempts me. Not sure if I'd bother with photo for double-sided boards.

>>1238844
>through-hole
disgusting

I want to invent a pen that allows you to literally pick up any material, draw a circuit on it, and then you'd just have to drill holes and solder shit.
You could even draw one circuit, put another layer of material on top and make more complex circuits with a smaller footprint.

>>1238774
>create a hand cranck generator
>>1238774
>put 2 motors in series for twice voltage
get even more voltage by using generators instead of motors

File: 1502355278652.gif (491KB, 226x226px) Image search: [Google]

491KB, 226x226px

>>1238863
>drilling holes
>2017

>>1238862
>I hear you can use an LCD screen with a UV light source to pattern photosensitized pc blanks

It's really shitty, unless you have a screen with an unusually high PPI. Consider that even cheap fabs offer 6-mil traces as standard. Just one pixel on an HD screen that's 16" wide is over 8 mils. Nevermind issues with contrast/leakage/etc.

Personally, I use photoresist ink and laser transparencies. Never had much luck with dry films; they keep coming off in the etch. Toner transfer is peasant tier. Well...actually it's near dumbass tier at this point. Cheap chinese photoresist is a trivial expense (especially when compared to the cost of the boards themselves + components), and, in my experience, is more reliable.

>>1238872
pls anon it's the only time i get to penetrate something

File: epaint.jpg (7KB, 275x183px) Image search: [Google]

7KB, 275x183px

>>1238863
conductive glue, but you can't solder it

>>1238875
>pre-sensitized boards from china for the same price as local substrates
That did not occur to us, dude. Now OP is slightly less of a faggot.

>>1238781
i get small boards (<2x2) from oshpark and larger boards from elecrow.

>>1238862
>>through-hole
>disgusting

Hey I'm etching this thing with a permanent marker and ruler, I don't have room to be picky. 2.54mm is narrow enough to make my head spin, I don't want to make it worse.

>>1238863
Not really paint but...
https://www.amazon.com/inch-yds-Copper-Foil-Tape/dp/B00EY44I42

>>1238866
its the only thing i have

>all these people shitting on through-hole for hobby projects

Are you a tech or an engineer? I want to create, not waste my time with the latest Chinese manufacturing technique to make my board ever so slightly smaller and prettier.

>>1238911
1206 and 50-mil SOIC aren't bad to work with, even by hand. If anything 1206 is almost a little clunky.

>>1239032
>fallacy of the excluded middle
See above. It is undeniable that you get access to a library of a lot more interesting, powerful, practical, and convenient parts with SMT than you ever will with through-hole. If you would rather waste your time clipping and soldering leads one by one, that's no skin off my back, but I'm a lot happier just taking hot air to the board. You might too, if you create anything that uses fine-pitch components because you literally cannot buy the function you want in DIP.

File: wiring pen.jpg (93KB, 768x512px) Image search: [Google]

93KB, 768x512px

>>1238863
http://www.roadrunnerelectronics.com/Wiring-Pencils

Like this?

>>1239056
>aren't bad to work with
Not with my chunky pen. I'll come back when I've got a 0.5mm pen and a microscope. I'm sticking to analogue and basic 7400 series digital so I don't mind in the slightest. Besides, through-hole looks better and you can't get SMD metal can packages that aren't crystals.

>>1239058
I was reading about wire wrap pens the other day, neat.

File: AVR.png (37KB, 1345x368px) Image search: [Google]

37KB, 1345x368px

>>1234666
>voltage regulators cause voltage spike

Is this true?

>>1239140
There are ac voltage regulators that automatically switch taps on a transformer in order to regulate voltage. It sounds reasonable.

File: wire-wrap.jpg (19KB, 263x250px) Image search: [Google]

19KB, 263x250px

>>1239120

that pic doesnt show wire-wrap. if you zoom in you can see all the connections are soldered. then they run the wires through mazes instead of using the shortest path, coz they're insane and have hours to kill.

real wire-wrap looks like this.

>>1239143
Can't they just shunt it with a TVS diode?

>>1239140
>Disconnect line-in
>Switch taps on the auto-transformer
>Reconnect line-in

That shouldn't create a nasty spike

>>1239151
Alright, what just happened to your line out?

>>1239153
The instantaneous current remains the same from the transformer's inductance so the voltage drop across it remains constant, no?

>>1239163
Disconnecting a tap of the transformer could create an inductive spike, transformers aren't ideal. But with some sort of Zener diode type device to shunt any peaking voltages to ground and/or good filter caps I can't see how this would ever be a problem since the switching times will always be so small. Though that's assuming you're rectifying the voltage, and I'm not sure what context is being spoken of. Rectifying diodes with caps across each AC out to absorb spikes with resistors to discharge them? If shunting the current prevents a high voltage spike, then it should work to some extent.

File: servo-buck-boost-7kW.jpg (75KB, 634x924px) Image search: [Google]

75KB, 634x924px

>>1239217
Switched tap is an outdated technique. Now buck-boost regulators are used, either servo or fully electronic. Variable transformers are make-before-break, hence the heat sink.

https://www.humblebundle.com/books/electronics-programming-make-books


worth it? I would like some easy-to-start idiot friendly books to get me started into electronics, Raspberry PI and arduino

>>1238764
You want the impedance angle in the stator coil that has the capacitor to be as close to 90deg as possible, though 80 is fine.

>>1239242
actually, the difference in angles between the two phases should be ~90, but since the primary coil only has an inductance the impedance angle of the coil with the capacitor can be lower than that.

File: 20170905_104438.jpg (1022KB, 3264x1836px) Image search: [Google]

1022KB, 3264x1836px

Is my ghetto arial going to help my radio? I'm trying to get the best reception for it as I can with irma incoming.
It's just spare electric wire from a ceiling fan, about three feet in total. I striped about 2.5 inches from one end and half an inch from the other, and attached the smaller end to the existing antenna on the radio

I know nothing about radios other that I've heard you can extend reception using metal

>>1239246
>with irma incoming
Who is 'irma'?
At that level of receiver it is best determined experimentally. The optimal length should be one quarter of the wavelength, which is about 75cm for the VHF broadcast band ("FM"). You do not need to remove the insulation at the end of the wire, just add enough to the telescopic part to arrive at 75cm and see if reception gets better. You could also push it fully in and connect 75cm with a crocodile clip or something like that.

>>1239251
Irma is a catefory four hurricane that's going to go up Florida dead-center in four days, there's no evacuating it.

Thanks for the tips, I noticed that even though I added the wire, reception only changed if I moved the original telescopic antenna itself.
I'll experiment a bit and report back.

>>1239240
Yes worth it

>>1239253
>>1239251
Opened up the unit, it appears the antenna is hard-soldered straight to a contact on the PCB. Could I solder my wire to the antenna?

File: 20170905_114120.jpg (2MB, 3264x1836px) Image search: [Google]

2MB, 3264x1836px

>>1239268
Ayy forgot the pic

>>1239253
Cat 5 now

>>1239272
Christ Almighty I'm gonna die

>>1239268
>solder my wire to the antenna?
I would first see whether a longer antenna helps much at all. The best way is to fully retract the whip and clip your wire antenna to the end. That way you can experiment with different lenghts. Be sure to have a good contact between wire and whip.

>>1239272
>>1239273
Why don't americans stop having hurricanes? It's like only them and south east asia who still get hit.

>weird grid voltage/freq
>weird unit system
>weird natural disasters
When will america become first world

File: sand_storm.jpg (862KB, 3000x2000px) Image search: [Google]

862KB, 3000x2000px

>>1239288
#NotAllAmericans
for example, in the midwest we get sand storms and tornadoes...

Where do you guys get small electronic components like sensors, chips and stuff? I need somewhere that's not expensive as fuck (like Conrad selling a barometer IC for 20€) and has worldwide shipping.

>>1239338
aliexpress.com

>>1239339
Yeah, that's what I already use. Got some pretty good stuff off it, and you can get some good stuff of the alibaba business market thing as well. I was mostly looking for other websites so I have alternatives when something is not available on ali.

>>1239338
>barometer IC for 20€
which one?

>>1239347
One of the only two, the other being like 40€

>>1239273
>strongest storm recorded in the atlantic
see you in the afterlife fellow floridian

>>1239377
>>1239273
Imma be fine, though
Florida west coast best coast

>>1239268
>Could I solder my wire to the antenna?
It won't do any better than just connecting the wire to the exposed antenna.
You should put it back together and quit fuckin with it while it still works.
I don't think you want to sit through Irma with a broken radio.

NEW BREAD

>>1239434
>>1239434
>>1239434

NEW BREAD

File: pressure-sensors.jpg (40KB, 560x516px) Image search: [Google]

40KB, 560x516px

>>1239338
>barometer IC
poor search term, use sensor -> pressure
you'll find many around 10€, Bosch, NXP, Freescale
Conrad.biz and Reichelt.com
1bar = 1e5 Pa = 100 kPa = 1000 hPa

>>1239226
It's still what's used in UPS devices since they reuse the transformer.