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



Thread archived.
You cannot reply anymore.



File: 1503385444366.jpg (68 KB, 848x635)
68 KB
68 KB JPG
bump limit reached on old thread >>1284905

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

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

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

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

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

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

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

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

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

>My circuit doesn't work. Halp?
Check wiring, soldering, part pinouts, and board artwork if applicable, then post schematic. Supply ALL relevant info and component values.when asking a question.
>>
File: 1507764330483.png (535 KB, 1716x1073)
535 KB
535 KB PNG
This thread's digits brought to you in part by the MAX971 ultra-low-power comparator with reference and E-Z hysteresis.

Thread pic dedicated to
>>1288541
>>
>>1288971
well for a start fuck you op for that image on a blue board.
>>
>>1289006
It's just some electrotherapy, don't know what you're getting your boners in a bind for.
>>
>>1289006

Shock treatments make you calm and docile.
>>
File: 1494824022591.jpg (49 KB, 800x533)
49 KB
49 KB JPG
>>1289006
Electronic, and not a sex toy or weapon, therefore topical.
>>
So dumb question:

If you charge a capacitor with a 9v battery, does it come out as 9 volts? I'm sure I'm not even asking a valid question.

I've read about capacitors but I'm clearly not thinking about it right.
>>
If I am new to the hobby, is it worth buying a misc lot of new components, or should I salvage some stuff first? I am looking at buying this:
https://www.ebay.ca/itm/161880646707
But I also have a rather large box of pcbs with plenty of stuff on them. Suggestions?
>>
>>1289129
yes it's worth buying a large variety of components
no that ebay listing is not worth it
buy from aliexpress, pay way less (and free shipping), and get more
>>
>>1289126

caps have various useful properties, but from a DC perspective, you can think of it as a rechargeable battery with a very small capacity. so, if you hook it up to 9Vdc, within a very short time, it'll charge to 9V and hold it for a few days. if you then hook it up to an led, it'll light for a second as it dumps its charge into it.

>>1289129

it's way better to have a ready source of new components. for example, if you're breadboarding, you often have to adjust a resistor value from 10K to 6.8K to 5.6K in order to get the result you want. if you're having to look thru PCBs and pulling out resistors, it becomes almost as tedious as reading the anti-tripfag complainers here on /diy/

on the other hand, almost every project you do will require one or more components which are not included in the kits, so you still have to order shit online and wait.
>>
>>1289135
>buy from aliexpress, pay way less

this will drive you mad. coz you wont know if your circuit is not working because of your own errors, or because the chink parts are fakes, or out-of-spec.
>>
>>1289135
>>1289137
>>1289141
Thanks for the advice. Will probably make a monthly "shopping list" for digikey (decent stuff and cheap shipping in Canada, everyone else is Ali or pricy) and make do with what I have for now.
>>
>>1289141
But that way you will learn how to troubleshoot stuff in no time!

I genuinely hate chinese and their national pride in making and selling fake stuff.

I bought few dozen of 2SC3355, half of the transistors had wrong pin-out (E-B-C instead of B-E-C). I got fake IXYS IGBT's, fake ST MOSFET's, fake el-caps, fake IC's (even fake TL072!).
>>
>>1289152
Here's your (You) for existing, come again!
>>
>>1289150
>digikey
you're going to spend 4x to 10x more than you would have spent on ali
>>
>>1289159
>you're going to spend 4x to 10x more than you would have spent on ali
Often you get what you pay far. A mate of mine bought shit ton of jumper wires, the wires and connectors were made of steel. No joking, they were magnetic. But yeah it was cheap.

Getting a bunch of resistors or low-voltage caps is OK while getting semiconductors could be risky (getting 2n2222/2n3904/2n5401/bc556 and stuff like is also OK, getting some RF or power electronics stuff is where you must be very careful).

Unless you have tools or time to build test benches I'd advise against getting ali-stuff for anything except packs of resistors/caps/popular small-signal transistors.
>>
>>1289159
>>digikey
newark.com often has nice prices on overstock items.
>>
>>1288971
god fucking damn it this shit will not solder.
I need to modify a wire, its this aluminum or zinc or whatever garbage. solder will not stick to this shit, help please
>>
Anyone had any particular chinkshit sellers who have sold them junk?

>>1289150
You should definitely order some R and C kits from aliexpress in the mean time, maybe some transistor assortments. Micros might end up cheaper from digikey, but note their minimum order requirements.

>>1289201
More and stronger flux? I'd seriously consider crimping.
>>
>>1289202
I might have to crimp but I really dont want to.
its on a computers power supply so Im wanting it to look somewhat clean
>>
>>1289201
nickel coated i think. need rosin flux and/or more heat.
>>
>>1289205
this cables are made for crimp. a lot of atx power suplies have crimped wires.
>>
>>1289206
Im using a weller 8200N 100/140 watt gun from the 1980s, its putting enough heat to start messing up the plastic coating on the wire.
I sorta got it to stick. I might just have to call it good enough and only make the necessary modifications.

>>1289210
inside the wires are soldered to the pcb.

thanks all of you for the advice
>>
>>1289168

MR-T, you are so full of shit.

How new are you to be surprised in finding magnetic leads & terminations. And yet you attempt to speak with any sort of authority? Unreal.

http://fiskalloy.com/product/c194/
>>
>>1289214
>http://fiskalloy.com/product/c194/
hahahah maybe too much heat. seriusly this is rare. i just soldered some of this with 40w iron and tin silver alloy
>>
>>1289218
I dont know. Im starting to think its a combination of aluminum being shit and the wires acting like a heatsink.
I seriously doubt 140w is too much.
is it really too much to ask for to not have a half dozen ide power connectors strung across your case like christmas lights
>>
>>1289218

copper-iron alloys are as common as all fuck. go test some through hole resistor leads with a magnet, faggot.
>>
>>1289222
Soldering guns are not fit for electronics work. Get rid of it.
>>
>>1289226
what makes them unfit?
>>
>>1289227
Too hot, burns off flux before it can act
Imprecise
Poor ergonomics
even the $5 fire starter from Radio Shack is a better choice
>>
>>1289231
>$5 fire starter from Radio Shack
god I cant imagine what that even is or how bad it was.
Ill try my $12 25 watt weller wood burner soldering iron later. I just cleaned up.
I hope that works out better, this is kind of advice that keeps me here, thanks.
>>
>>1289232
Basically a 30W weller wood burner clone, maybe they were more like $10 after I ever bought one
It worked, for the good old .100" through hole stuff
Might or might not work well with lead-free solder which needs a rather higher temperature to melt, which could also be the problem with the soldering gun
>>
I am late to this soldering trouble party. If you have a wire that has corroded up behind it insulation, it is going to be a bugger to get the solder to wet. You can/should try a more active flux and bump temperatures up a bit. If no love try using some of the black metal sandpaper maybe 150 grit - DON'T use the normal brownish wood sandpaper - the abrasive grains will embed into the copper and you will never get solder to wet after that. pinch the stripped wire hard (even stranded) in say a 1" wide strip of the black stuff between your fingers and pull it out with the other hand. Do that a dozen times to break up the oxide layer. clean the wire end with some alcohol, flux and trying soldering again.

That should work.
>>
>>1289233
Im using 60/40 solder. oh wtf.. no I have lead free solder. didnt know that. my old solder was 60/40
you do know Im not doing through hole, but am instead trying to solder 18 gauge wires that are a few feet long and connected into a huge bundle in a pcb that then go back out to feet of more wire and is probably acting like a heatsink, right?
>>
>>1289169
>>1289168
>>1289152
I want tripcunts to leave.
>>
>>1289257
I was tempted to call this "apply power to the tripfags" edition, for about a second
>>
File: I-FUCKING-HATE-CHINESE.jpg (122 KB, 1440x877)
122 KB
122 KB JPG
>>1289216
>MR-T, you are so full of shit.
>How new are you to be surprised in finding magnetic leads & terminations. And yet you attempt to speak with any sort of authority? Unreal.

Unfortunately I'm not that new, I dealt with magnetic leads (diodes, old germanium transistors, elkos) in very early 90s.

I only once got sanken transistor made of steel: it has steel head spreader and leads, I broke it with a hammer and the silicon chip was way smaller than it should be. In few days the heat spreader starter to rust.

Chinesium wires are made of copper clad steel, connectors are made if tin (for real, they rust). If you scrape copper from wires you'll see whitish shine and in few days you'll see rust. How new are you?

I'm attaching a picture if some fakes I have on hand at the moment. ST fets - one is fake and one is real, IR IGBT is used and has new terminals welded, all of the other parts are sanded and have new marking.
>>
>>1289281
And there is no such transistor as FQAF11NK90C, there is FQAF11N90C. What I have is a BJT NPN and not a MOSFET. Probably a typo.
>>
>>1289281
>>1289282
We don't care. Drop the fucking trip.
>>
>>1289281
>Unfortunately I'm not that new
Ahh, so you're just a stubborn old cunt who is not only bored by us kids, but is too pig headed to do like everyone else. Got it.

Tell us more about how much of a faggot you are.
>>
>>1289265
I don't care. It's the content that counts.
Triggered children we have enough.
>>
I still fucking hate breadboards.
>>
>>1289201

Dip them in nitric acid then wash it off in distilled water, then apply flux as before.
>>
>>1289129

Advice 1: Leave them on the PCB until you're ready to use them.
• You can get rare/old stuff from some of those PCB.
• If it's from a good manufacturer, those components will likely be the real deal.

Advice 2: Don't buy grabbags full of crap.
• Figure out what you're going to build, and buy those components, but when you do:
- get them from a reputable source.
- buy extras.
- go "up" in wattage and voltage temperature, speed, and case, ...
(i.e. get 1/2 watt resistors instead of 1/4, 1N5408s instead of 1N4001, 105 deg caps, metal-can 2222s and TO-3 3055s, etc, etc....
>>
>>1289282

What, are you kidding me?

Just because you can't find a data sheet for it doesn't mean that it doesn't exist! All the semiconductor mfgs will make you a custom part if you order enough of them, I have many such parts (i.e. dc/dc modules) that have no "retail" part number; but nontheless exist.

Watch (and learn1) what transistors he's using:
https://www.youtube.com/watch?v=rpGOKGrcpAk
>>
So when a transistor is in saturation, do I just completely disregard beta?

And specifically for pic related, do my equations become solely based off of Kirchoff's laws?
>I_c = (V_cc - V_ce) / R_Load
>I_b = (Vbb - V_be) / R_b
>>
>>1289478
more specifically
>I_c = (V_cc - V_ce_sat) / R_Load
>>
File: 2222sat.png (47 KB, 805x378)
47 KB
47 KB PNG
>>1289478
>..disregard beta?
No, it is a documented design parameter.
>>
>>1289201
Use zinc based flux, that shit will let you even wooden bricks (E.g. LW25 WELLER). Apply some, let it work for few seconds, solder.

Just avoid it like a plague and clean the surfaces afterwards - very toxic and can corrode PCBs/wires if left uncleaned.
>>
File: flasher.png (13 KB, 819x456)
13 KB
13 KB PNG
In last thread I made a simulation of this traffic barrier flasher circuit with Logisim. The simulation worked fine, but the JK-flip-flop had spurious signals at its output. Eventually I managed to solve the problem by replacing the 74F374 with a CMOS part (M74HC164 shift register) and grounding all of the unused input pins of the CD4027.
>>
File: output.webm (1.17 MB, 640x480)
1.17 MB
1.17 MB WEBM
>>1289558
And here is the circuit in operation. Filmed with a potato.
>>
>>1289281
>connectors are made if tin (for real, they rust)
>tin
>rust
No.

>>1289463
>metal-can 2222s
Yes.

>>1289478
Yes, otherwise you'll be pulling more current through the collector and the voltage across the load will be greater than Vcc. That is to say you disregard the beta being a constant value.
>>
>>1289558

Brother, this is super awesome. You are close to the level of understanding to get into fpgas (if that interests you).
>>
>>1289558
>IEC logic gates
Into the trash it goes.
>>
>>1289564
>No.
Chinese do sell super-thin steel wires and they come with steel connectors. They do that because steel costs way less than copper.

Most DIY'ers that get them are just too dumb to realize that.

> I found another breadboard wire kit from an Elenco supplier but the breadboard wires are probably steel so they aren't copper.
http://forums.parallax.com/discussion/122934/good-breadboard-gauge-wire

Peps I urge you to get a magnet and check your bread board jumper wires & connectors. If they are straight from china they maybe made of steel.
>>
File: jw.jpg (251 KB, 800x609)
251 KB
251 KB JPG
>>1289640
that kind of wires
>>
>>1289469
>Just because you can't find a data sheet for it doesn't mean that it doesn't exist! All the semiconductor mfgs will make you a custom part if you order enough of them

I love that attitude. Show fake chinese part which is marked as a typical FET and has a typo (ST has "NK60" mosfets, fairchild doesn't) and yet it's a BJT. There are no datasheets for it.
>NO I DON'T BELIEVE IT, YOU ARE WRONG, IT'S A CUSTOM PART.
Perfect example of Danning-Kruger effect.
>>
As I understand it BJTs can be used as zeners, so if I needed a very rough 60v zener I could use a 2n3904 with C as the cathode and BE as the anode. Are there any caveats to this other than power dissipation?
>>
>>1289651
>As I understand it BJTs can be used as zeners, so if I needed a very rough 60v zener I could use a 2n3904 with C as the cathode and BE as the anode. Are there any caveats to this other than power dissipation?
I saw few examples when people actually used E-B junction as a low-power / low-voltage zeners but never C-B junction. It should work. Caveats probably low power dissipation, temperature instability, non-controlled break down voltage may result is big Vbr difference across the samples.
>>
>>1289463
>don't buy grab bags full of crap
Correct. Instead buy sample kits so you have a decent stock of values for resistors/caps/basic transistors. It's hard for even unreputable sources to fuck such basicness up.
>go "up" in wattage
Sometimes.
Do NOT get 1/2W *instead of* 1/4W unless you're a hipster or have very very poor eyesight. If you need 1/2W you're probably working with much higher voltages than a noob would ordinarily be working with. Anyway you can make a 1/2W resistor out of two 1/4Ws, every time.
>voltage temperature
Yes, voltage and temperature derating is always good and increases component life.
>speed
Sometimes.
>, and case
No. I can't even think of when the last time I saw a TO-3 was. TO-220s conduct heat better, are cheaper to buy, mount, and cool, and don't have the live-case problem.
Presumably, anon's going to want to commit a project to perfboard someday, and draw from the experimental stock to do so. Gigantic components are not helpful toward this end.
>>
>>1289201
I have some dupont wires that are impossible to solder as well
solder will bead up on it and fall off, even at 480c
>>
>>1289667
> hipsters
> to-3

To-3's are perfect for 2nd gen hipsters, 1st gen hipsters have goddamn TUBES in their headphone preamps with glass windows so you can see their majesty. Prob why we mount to-3s on the OUTSIDE of the case, lol!
>>
>>1289650
>Perfect example of Danning-Kruger effect.

haha. talk about the pot calling the kettle black.

Let me guess: you were kicked off whatever other forums you 'blessed' with your presence and 'intellect'. Then you found your way here, not the kind of place that bans people. But the anon format here does not give you the sort of personal validation you need/crave. Hence the Trip.
>>
>>1289737
I wasn't banned anywhere since I abide the rules and contribute a lot. Neither I was kicked for whatever reasons.
I'm here just to practice English, learn youngster slang and entertain myself while helping others.
What makes me genuinely wonder is why the hell anons on /diy/ are so annoyed by trip codes. If you are here to share information then go ahead, ask questions, get answers, give advices, no problems. If you are here to shit post then again, literally trip codes give you zero problems as well. What do we have left? Some skewed anons with defective perception of reality where any poster with a name or a code tag other than "Anonymous" is pure evil? Nice.
>>
File: RBv5Z[1].png (63 KB, 1125x681)
63 KB
63 KB PNG
>>1289742
>>
>>1289640
TIN DOESN'T RUST
>>
>>1289742
i'm not him and haven't bitched about your trip. the reason anons hate trips is because when one shows up in a comfy little thread like this, half the thread inevitably becomes about him. it doesn't matter whether or not that's your fault or the fault of the people replying to you. if you care about the thread, stop using the trip. if you don't care about the thread then you've got the hate coming.
>>
>>1289746
Essentially this. But tripcodes are in the first place to be used when your identity matters. By using them here you're implying your identity is important, so no amount of "i'm not an attention whore, and this isn't hurting anyone" is going to change the fact you're wearing a 16 foot name tag over your midriff.
>>
>>1289747
femanon here, I'm wondering how this 'tripcode' thing works.
Can I get one? Can someone make it so everyone will know I'm a girl?
That way I'll get the special attention I'm accustomed to for being of the fairer sex.
Thanks, guys <3 -smooch!-
>>
>>1289763
/ohm/ loves girls, you just have to post your dick as proof.
>>
>>1289764
It's true though.
>>
>>1289650
> Perfect example of Danning-Kruger effect.

BTW, It's “Dunning” with a ‘u’
Perfect example of the Dunning-Kruger effect.

I didn't go over the original image like zapruder film, I just responded to the one post.

Here's some interesting trivia for ya:

Sometimes the existence of “fakes” is the original manufacturer's own fault. Some of them actually sell-off (in bulk) the die material and/or finished parts that failed every bin test.

In the latter case, they may laser-etch the failed part with an indicator that it failed, so that has to be sanded off and re-printed.

What did they think the Chinese were going to do with it?
>>
>>1289742
since you seem genuinely new and not just a cunt, so Ill answer your question.
Tripfags have always been hated. this is an anonymous image board, not reddit. Tripcodes are made for throw away verification for something like verifying an anon actually delivered, or if an anon needs to follow up with something later on, it can be used to show that someone else isnt just making a troll thread full of bs.

attention seeking faggots like to use trip codes because they want a username and the credit associated with their posting. its the same kind of "upvote, comment, like, and subscribe" bullshit that has made the internet so miserable.
If you want to attention whore or rep whore then go to reddit.
If like the culture here, then assimilate.
>>
>>1289742

Are you Ronery, Mr.-T?

https://www.youtube.com/watch?v=xh_9QhRzJEs
>>
>>1289787
Fucking this. Go look at >>>/g//bsg if you want a prime example. Never mind the fact that here on /diy/, there are literally no tripfags apart from this guy, which I feel is to the credit of the board. Identity is vehemently opposed, no matter how good your information is. With no reason to "maintain a good name", all thats left is information. If it's good, congratulations, you'll be thanked, and you will know you have genuinely helped someone. If it's shit, you'l be told so and you can slink off and try again. Like this guy says, if you feel a need to attach an identity to your information, your motivation is flawed, and you don't belong here.
>>
>>1289815
I think that if there was anyone legendary enough to just always have a tripcode it would be bunkerbro, and even he doesn't; his images speak for themselves. But even then it wouldn't do to see bunkerbro shitposting on /bant/ with his tripcode, it really should be reserved only to topical posts.

Or simplified, use a tripcode when your actions have already established an identity for yourself, don't use a tripcode to establish that identity in the first place.
>>
>>1289837
>his images speak for themselves
Exactly. See also; 6 axis robot guy. Again, information over identity.
>>
Quick question about Atmel ATMEGA386 with pre-programmed bootloader: if I use an AVR ISPMKII programmer to write my .hex to flash memory (avrdude invoked with -e switch) I effectively get rid of the bootloader right? Is there anything else I should do to make the chip "factory default" (excuse the poor wording, but I think you get the gist of it)? Like maybe rewrite the fuse bits or something?

I just want to use it with the basic avr-gcc toolchain to learn.
>>
>>1289841
*328

Don't know why, but I had an Intel moment there.
>>
>>1289841
-e should be good, you'll probably want to change the fuse bits when programming it next time, according to that project's particular needs (clock sources, watchdogs, etc)
>>
File: asd1.png (23 KB, 444x200)
23 KB
23 KB PNG
I am a retard and forgot the pic in the firstplace.
again:
today I build this machine with only logic gates/flip flops. webm coming
>>
File: machine.MOV.webm (2.49 MB, 1280x720)
2.49 MB
2.49 MB WEBM
>>1289995
top right green is x
blue led is clock
other green is just button indicator
red is current state (left 2^0 right 2^1)
>>
File: DSC_0032.jpg (716 KB, 1548x2064)
716 KB
716 KB JPG
>>1289558
I returned to using the 74F374 as a shift register because it has a useful Output Enable pin for PWM'ing the outputs.
I added another 74F374 flip-flop, light sensor (LDR) and a comparator IC (MAX921).

The MAX921 output turns on when the resistance of the LDR gets higher in the dark. The comparator output is connected to one of the inputs of the Schmitt-trigger NAND (CD4093), turning on a ~200 Hz astable oscillator. CD4093 oscillator output drives the Output Enable pin of the second 74F374.
>>
>>1290065
What is the second 374 for? You could also use the LDR to gradually pwm the LEDs so that brightness follows ambient light.
>>
>>1289996
Newb question here. I know where you can buy these chips, but how do you figure when you need them? And where can you find a good list of these different chips and what they do?
>>
>>1290124
First you know what you want. Then you determine the functions you need to achieve that, then you draw (and calculate) the circuit and look for parts that offer these functions. Electronics is design, not shopping.
>>
>>1290124
discrete logic chips fall into a few types, and then there's a selection of chips with slightly different details among those types.
>shift registers
>logic gates (AND/OR/NAND/etc)
>multiplexers
>counters
>etc
so you just have to learn what "shift register" does and then when you need something that stores data or converts parallel/serial then you go and look at the different options for shift registers and see if any fit your requirements.

common discrete logic families can be found in those lists. alternatively you can search digikey's shift registers category for things not included there.
https://en.wikipedia.org/wiki/List_of_4000_series_integrated_circuits
https://en.wikipedia.org/wiki/List_of_7400_series_integrated_circuits
>>
>>1290124
You can read whatever generic tutorial on logic to get an idea of the available basic blocks and methods. In the olden times you'd then take a thick TTL databook to see what functions are actually available and pick convenient-looking ones. For example, your tutorial might suggest that to decode 3 inputs into 8 outputs you'd need a bunch of gates, but in practice you'd pick one IC which does the job.
It's 2017 now, though. Use relevant manufacturer's (like TI or Nexperia) parametric search. Or check Wikipedia's lists.

https://en.wikipedia.org/wiki/List_of_7400_series_integrated_circuits
>>
>>1290124
Read >>1288971
>>
>>1290140
And where does it address his actual question?
>>
>>1290140
Catalogs are lists, not necessarily good lists.

>>1290065
You could insert a decent power transistor to turn the top-side supply for all the LEDs on and off for PWM. It's like a multiplexed matrix, with only one row. At 4mA per LED, a 2N3904 could do that..
>>
>>1289559
>output.webm
Why are webms broken on /diy/
>>
>>1289559
cool little light show. can u change your chase patterns?
>>
>>1290255
works4me
have you tried resoldering all the joints
>>
File: leds.jpg (25 KB, 307x523)
25 KB
25 KB JPG
>>1290166
Low side pwm would be easier because anodes go to individual register outputs. Breadboard is full though.
>>
>>1290336
>full
But you can now lose one of the F374s and just switch the other side of the LEDs. You generally get better drive out of the low side of TTL outputs, so a common high-side would be a better design.
>>
>>1290339
True, TTL likes to sink.
>>
What tools and essentials should a good hobbyist electronics setup have? I just snagged a nice Pelican case(it's even Fluke branded, which is perfect), and I'm thinking about getting the foam and just putting all my electronics stuff in it, since I don't keep a station setup for it. Got a decent Chinese solder station, Engineer solder pump, desoldering braids, sets of silicone and polyurethane? sleeved wires, Kester RMA solder and flux pen, a DMM, going to pick up or make one of the tiny portable digital oscilloscopes, and some prototyping board/breadboard and the jumpers for it. Anything I'm missing before I start cutting foam?
>>
should I buy a special chip to measure voltage across a current shunt?

It's just measuring 0-100mV with an arduino, which is super easy to build with an opamp. But the measurements around 0 are the most important, I'm worried about it not being very linear at the edge of the input range. how do I guarantee accuracy near 0?

Also, why are all of the current sense amps built for high-side measuring? low-side would be cheaper/easier, right?
>>
>>1290406
Whatever you actually use. Beyond those very basics, tool sets will depend a lot on what you're making or repairing.
>wire
No plain old hookup wire? Clip leads?
>RMA
Consider flux remover.
>portable DSO
If you're only doing audio-frequency work this might be worth it, but otherwise you're better off getting a Rigol.

>>1290408
Op amps will be specified for which rails are available at the input and output. The LM358-type op amps are fine at the negative rail for input and output, but don't work so well near the positive rail. Check the datasheets.
>why are all of the current sense amps built for high-side measuring
Because low-side measuring is cheaper and easier, but sometimes you can't measure on the low side.
>>
>>1290410
>No plain old hookup wire? Clip leads?
>>1290406
>sets of silicone and polyurethane? sleeved wires
Got wire, and a buddy just got me a damaged 87-V with all the accessories, so I got a spare thermocouple, set of leads, and a set of clip leads. Meter works too, except for mA measurements. I think I'll look into getting some more variety of leads before I start cutting foam though.
>>
File: foo.png (2 KB, 242x245)
2 KB
2 KB PNG
does it make a difference if I arrange cells like in top or in bottom?
>>
>>1290455
You can use a normal protection board for the bottom setup (two parallel cells can be treated as one) and it balances better.
>>
>>1290457
They're just alkalines, not gonna use any protection.
>>
>>1290455

the top arrangement is better, coz it makes matching easier, by lessening the odds of having a weak cell paralleled with a strong cell.
>>
File: sonoff-program-3.jpg (205 KB, 1280x751)
205 KB
205 KB JPG
I've never really dabbled in electronics before but how difficult would it be to hook pic related up to my lightswitch?
I'm going to buy it from China and I live in the UK so my light-switch might not work.
From what I can tell it should be easy enough to hook up my bedside lamp but light-switches in the wall might be a lot more difficult.
>>
>>1290620
First things first: what is that? Some kind of timer?
>>
>>1290637
It's a sonoff wireless switch, you attach it to cables and it allows you to control the outlet over the internet
>>
>>1290620
You will not necessarily have a neutral at the switch either or the lamp itself.
So it depends on the wiring in your house, either the mains goes to each switch then a neutral and switched line to the light or the mains goes to each light and a drop of line and switched line to the switch. So you see in the second setup you don't have a neutral at the switch so it needs to be wired in at the light fitting.
It is absolutely possible but screw terminals are no way maintenance free so really you shouldn't go burying it in walls or under floors.

Oh, if your house is old out might be all switches and lamps are wired back to a huge junction box somewhere.
>>
>>1290620
If you buy a 240V version you should be fine. Wiring it into light switches in the wall isn't technically legal if it isn't inspected by an electrician afterwards, if that gives you an idea of how tough it could be, and they wouldn't allow some un-certified import electronic doodad plugged in there in the first place. Wiring it into a lamp cable will be a piece of cake, though still probably worth a house-fire or two in the eyes of an electrician. Perhaps look for one that screws into a light socket that the bulb screws below, or one that you plug your lamp plug into so you don't have to cut its cable.
>>
What is the smallest project you have ever built /ohm/?
>>
File: asdass5.jpg (1.03 MB, 1920x1020)
1.03 MB
1.03 MB JPG
>>1290733
this, but i still haven't worked up the will to learn to program silicon labs ucs and it needs to be cleaned again. it was partly just to see how small i could assemble things. i think i swallowed a few resistors using a needle and suction tube to place them.
>>
File: 1507256091344.png (221 KB, 560x560)
221 KB
221 KB PNG
>>1290750
don't eat resistors
are those two CSPs I spy there?
>>
File: 1.jpg (17 KB, 555x318)
17 KB
17 KB JPG
>>1289745
>TIN DOESN'T RUST
Tin or whatever plating on steel connectors wears out, exposes steel and it rust.
I scratched the plating and then poured tap water on it. 8-10 hour and you get rusty connector.
>>
>>1290902
>connectors are made if tin
That isn't how I'd articulate that component pins are tin-plated iron at least. But all my resistors from my budget "ten of everything" pack are also tin-plated iron. It's not really a worry for most electronics, provided you solder at a nice low temperature, but I wouldn't like to see that on a current shunt.
>>
>>1290922
I helped a friend of mine to troubleshoot issues with his arduino-based project as he was really new with EE. Servos and arduino/mcu were unstable and there were shit ton of various symptoms and glitches.
The root cause was these damned jumper cables (high voltage drop at low current and intermittent contacts). I took few as example of chinese craftsmanship, you know it's not that common to see thin steel wires and steel connectors.
>>
File: l.png (780 KB, 1267x871)
780 KB
780 KB PNG
I've been reading this old CMOS logic data book recently:

https://archive.org/details/RcaCmosI.c.Databook1983

Are discrete logic ics still used in modern electronics (outside of hobby projects)?
>>
what do people mean by "buy in bulk"?
Like, buy a box of 10,000 resistors or buy the "Beginner Electronics Kits 100 Components in One" kind of kits, or what?
>>
>>1290944
>>1290902
Oh look, his majesty is back. Die in a fucking fire you old tripfag cunt, you don't belong here.
>>
>>1291329
One-word reply would be "no".
The age of filling large boards with discrete logic is long gone. Nowadays discrete logic is mainly used as glue and even as glue, the amount is much smaller now than it was, say, two decades ago. It's no surprise all the logic chip makers now offer singly-packaged gates and such.
Of course, there are always exceptions. For example, the design might be so simple that discrete logic is a cost-effective option.
>>
>>1291329
There are a few parts in the 74xx and 40xx families that remain relevant. You'll still occasionally see analog switches of the 4066ish flavor, the odd 74xx595 or other mux/demux for microcontroller I/O expansion or SPI chip selects, some level shifters, bus drivers, that sort of thing. With microcontrollers being equivalent or superior at just about everything these days, especially design cycle time, simple logic gates are rare.

>>1291350
Context-dependent. Could mean anything from buying in the manufacturer package quantity (box/reel) to loose parts by weight.
>>
>>1291368
Really I'm just gonna make effects boxes for my synthesizers, and everyone and everything says to "buy in bulk"

Unfortunately I don't know the most common specific components that go into these things so I don't know what all to buy
>>
So if potentiometers are variable resistors, then what is a variable capacitor?
>>
>>1291384
>potentiometers are variable resistors
wrong
>>
>>1291388
>>potentiometers are variable resistors
>wrong
wrong
>>
>>1291384
Potentiometer has three terminals, variable resistor has two terminals. Variable ratio does not mean variable value.
>>
>>1291402
>>1291388
so now that the pedants have waxes eloquent would someone care to answer my question
>>
>>1290944
Jumpers would face problems because of the friction taken to insert them, but soldering onto a tinned iron lead would probably be fine.

>>1291406
They're just called variable capacitors, or varicaps for short if you want. They usually have pretty low capacitances, tens to hundreds of pF for the small ones, so if you're thinking of using them for anything other than a tuned LC circuit or other HF use then you're probably out of luck.
>>
>>1291406
if you want one that works like a potentiometer:
differential variable capacitor
>>
>>1291406
>>1291413
If you want one with higher values you can split the line with a potentiometer feeding two different capacitors in parallel, though your Q factor will be terrible. Still, if you're buffering your signal anyway, it could prove useful if you don't want to go digital.

I've never tried it or seen it though, so use at your own peril.
>>
>>1291374
For basic items like resistors and caps, I suggest buying a small stock of assortments and replenishing that stock as you run low. For transistors, buy some reasonable assortments plus 10- or 20-packs of the few oddball types you'll need. For parts that are non-dirt-cheap like polypro caps, specialized like ICs or vactrols, or most often used in only one or two types/values like voltage regulators or jacks, skip the assortments and just buy however many of those you think you'll use in a year or two. It boils down to budget vs. plans.
>>
>>1289787
I think at this point we can safely assume he does in fact want to rep whore, and should be treated as such. Nice try I guess.

>>1290944
>>1290902
Just fuck off man, you have shown your colours, you're not welcome here, and you rightly deserve the hate should you choose to continue.
>>
File: 1509733120317.jpg (90 KB, 714x681)
90 KB
90 KB JPG
So I'm working on a project for uni that will require both Arduino and Raspberry Pi. The pi will run a GUI that sends instructions to the arduino, while the microcontroller will handle the sensors and components and feed the information back to the pi (which I imagine will be done via serial port.)

Question is, what kind of pi and arduino do I need? Lot of people seem to just go with arduino uno and the Pi B+, but I'm wondering if I can get away with running all this off of a Zero W if I just power them separately. I mean I can certainly spring for both since the budget allows it, but if I Could get away with less somewhere I wouldn't mind it
>>
>>1291573
better just stick to the known quantity
>>
File: file.png (56 KB, 1894x1021)
56 KB
56 KB PNG
Does anyone know why my CMOS crystal oscillator isn't amplifying the oscillations? They die out after the excitation pulse. I've tried with and without R1 according to www.electronics-tutorials dot ws/oscillator/crystal.html, with the component values from the same page.
>>
>>1291579
the inverter in the original schematic has Schmitt-trigger inputs. your unbuffered inverter does not.
>>
>>1291582
I know. It shouldn't make a difference. https://electronics.stackexchange.com/questions/218142/using-cmos-schmitt-trigger-inverters-in-quartz-crystal-oscillator-circuit
This page says it makes more sense to not use a Schmitt-triggered inverter. That stage is only providing gain and 180 degrees.

Is it possible the power is bleeding out through C10 and the voltage source?
>>
>>1291579
might also try adjusting R3 to see if you can get more gain out of that inverter you made there. check the Vgs(th) of the FETs too and how much current they're passing.
that voltage source connected isn't helping maintain loop gain by staying connected, to be sure. maybe an open-drain buffer arrangement would be better.
>>
>>1291573
An Arduino Nano and Pi Zero should work fine, unless there are features you know you'll need that don't exist on those platforms. You could run a GUI with an Arduino alone, though having an actual OS helps things if the GUI needs to be complicated.
>>
>>1291579
Use real FETs for that, not just their default model.
>>
>>1291599
Also there are crystals in spice's directory, spelt with the shit spelling of "xtal".
>>
File: file.png (67 KB, 1920x1040)
67 KB
67 KB PNG
>>1291599
Any recommendations for suitable FETs? I tried using AO6407 and AO6408 and got 45A drain currents.

>>1291587
Lower R3 should result in higher loop gain, right?
>>
>>1291587
Meant to say that I changed it so that the pulse source is disconnected by a 1GOhm resistor.
>>
>>1291602
>45A drain currents
yeah that'll oscillate fine :^)
The FETs in an actual CMOS gate are much much smaller than that and will limit current on their own.
SI4558? DMC3032? If you don't have those, you could try pushing up the Vgs(th) on the ones you're using to about 2V or so. AO3400/AO3401 might also be a good pair of FETs. I'm tempted to suggest putting 50 ohm resistors on the drains to represent output impedance. Those are small relative to what you're driving.
>>
>>1291545
>I think at this point we can safely assume he does in fact want to rep whore, and should be treated as such.
yea I figured that out when he continued to be a raging faggot with posts such as "tin can rust because I damaged this iron connector by removing all of the tin from an area and the iron rusted"
>>
>>1291621
I'm using a BSS84 and BSP89, with threshold voltages of -2.1 and 1.6V. Still not enough loop gain. I tried putting on 50 ohm resistors and it didn't make any difference.

I'll probably just use a CMOS inverter IC model if I can find one. Got any recommendations?
>>
>>1291657
did you try upping the big resistor to 10Mohm and smaller load caps?
the 74LVC(1G)(U)04 is the recommended choice for a crystal oscillator. try to use a U version if you can. maybe your sim doesn't have 1G logic. LVC is probably the only family you'd try to use for a 10MHz osc at 5V.
>>
File: xtal-osc-cmos-inverter.png (78 KB, 1440x900)
78 KB
78 KB PNG
>>1291657
>. Got any recommendations?
What is your goal?
>>
File: IMG_1217.jpg (1.11 MB, 3264x2448)
1.11 MB
1.11 MB JPG
Alright, I've got a 2-wire brushless DC 12V fan I need to replace (0.19A), and the only ones I can find have three times the current rating (0.55A). Can I just put a resistor in series and expect it to run fine (putting the resistor inline with the airflow), or do they not tend to run well on 1/2 the voltage?

Running the thing with PWM probably won't work since it's brushless, but a small step-down module could be an option provided I can fit it in somewhere, perhaps a capacitive one. But I'd rather just rewind the motor or just try to salvage the stator itself. On that note, any idea how to get the PCB+stator off one of these?

Unless there are better places to be looking than AliExpress, of course.
>>
>>1291579
>Does anyone know why my CMOS crystal oscillator isn't amplifying the oscillations?
Gain is too low.
>>
File: cmos-buffer-gain.png (76 KB, 1440x900)
76 KB
76 KB PNG
>>1291715
>>
>>1291713
>resistor
Not really.
>small step-down module
Quite possibly. I've seen a very small step-down module on ali, but I doubt it'll fit inside the hub. If there is a corner without a screw hole you might be able to slip a converter board in there.
You sure you haven't got the extra 350mA to spare?
>>
>>1291713
You'll be fine either way. You can drop voltage a bit using a resistor or use PWM, just don't go crazy with PWM frequency.
>>
>>1291713
You sure your supply can't handle 0.55A?
>>
PS. I literally cannot stop sucking cock. Maybe if I suck enough you'll all start loving me?
>>
>>1291722
>>1291717
>>1291715
>>1291710
Here ya go you desperate old shitfuck. You clearly have not noticed that everyone is ignoring your replies, so have some (you)'s on me.

Now fuck the fuck off already.
>>
>>1289469
To further add to this, I swiped a gif from Wasatonic's web page. Much of the info in that shelf of data books may be gone forever save a few select bookcases. The manufacturer certainly has little reason to scan every datasheet they ever produced into pdf. And even if certain parts are catalogued into a database, it may be proprietary and ungoogleable.
>>
File: E213+08A+GIF.gif (1.6 MB, 1280x960)
1.6 MB
1.6 MB GIF
>>1291737
Oh yeah, the fucking gif.
>>
Just overspent on nice LED strips to /diy/ lighting in a few rooms, is there anything I should know or look for as far as power supplies and dimmers go? Or should I just buy whatever meets the wattage requirements (300+)?
>>
>>1291719
>>1291728
It might well be able to, but since it's not to spec and it's going to be fitting into a $800 induction stove I don't want to risk it. There's also a potentially unnamed fuse on the PSU board, which I don't want to blow.

>>1291722
I guess PWM could work if the frequency is significantly lower than that of the fan, but lower frequency = bigger filter caps. I'll see what I can fit into a 555-based PCB and see if there's room to shove one in. It's just going to be a 555, a trimpot, a transistor, a capacitor, and a bunch of passives anyway. I could also measure that it's getting the same airflow/power as the other fan with a straw in a glass of water, which could be interesting to try.

The 555 PCB will be smaller than a step-down module anyway, which helps if it works.
>>
>>1290656
Why are there no wireless outlets? As in, not an adapter you plug in, but an outlet that is connected to the actual hot neutral and ground wires so that it is always an iot device? Seems like it would be the future of "smart homes", particularly since the adapters are often quite bulky.
>>
File: dcap.jpg (16 KB, 655x191)
16 KB
16 KB JPG
>>1291415
yup
>>
File: 1481126114983.jpg (139 KB, 640x640)
139 KB
139 KB JPG
>>1291744
>The 555 PCB will be smaller than a step-down module anyway
You sure about that?
>>
File: 1512396807408.jpg (6 KB, 230x300)
6 KB
6 KB JPG
>>1291747
you gotta learn to google more
>>
>>1291743
I am also interested in this. Bought strip lights I still haven't installed because I got hung up on whether I wanted the cheap chinesium dimmers from Amazon. Or whether I should buy 12VDC PS units from some place like Digi-Key. Are there any industry standards yet, or kind of wild west?
>>
File: smart-outlet.jpg (10 KB, 500x500)
10 KB
10 KB JPG
>>1291747
The Ultimate Havoc: Hacking 'Smart' Homes for Fun & Giggles.
>>
>>1291753
Holy damn those are tiny! But I don’t have one lying about so I’ll go with the PWM for now. Still, those are great looking things to have in a parts-bin.
>>
>>1291783
It's not even just step-down.
These modules also do step-up too.
>>
>>1291786
>These modules also do step-up too.
Different modules, none does both.
SEPIC requires two coils and is larger.
>>
>>1291792
https://www.aliexpress.com/item/5pcs-mini-1-8V-3V-3-7V-5V-to-3-3V-Boost-Buck-Low-Noise-Regulated/32365767349.html ?
>>
File: ytrr1ew89G8[1].jpg (262 KB, 1366x768)
262 KB
262 KB JPG
Hi. My laptop on pic. Is it Gpu or vram problem?
How to know?
>>
>>1291713
>Can I just put a resistor in series
Why don't you try it?
https://www.google.com/search?q=fan+12V+0.2A
>>
>>1291794
That is a charge pump module for max 0.1A. Not usable to drive a 12V fan at lower speed. "These modules" shown at >>1291753 have a coil and 1A output and the output voltage can be adjusted by changing a resistor if need be.
>>
File: IOUT_0A.jpg (17 KB, 340x340)
17 KB
17 KB JPG
>>1291825
Clever deception though. Boost with voltage limiter marketed as boost-buck. No wonder there is no load characteristic.
>>
>>1289815
Or how about everyone ignore the trip codes and just read the posts? Seriously why does anyone fucking care whether someone uses a trip code?
>>
I have all these components. What should I build?
>>
>>1291921
>What should I build?
Whatever you want, e.g. nothing.
>>
>>1291921
Build a traffic light controller
>>
>>1291915
This is an anonymous image board
That's what it's for, that's how it works.
Names and trips are only really for emergencies otherwise its not anonymous.
If you don't like it then fuck off to one of the thousands of forums where people jerk each other off and get your ego stroked I guess
>>
>>1291800
It needs to be a 9733 fan.
>>
>>1291968
>traffic light controller
that's a challenge
>>
>>1291921
Something audio or otherwise analogue based, since you've got all those capacitors, inductors, op-amps, pots, and a speaker and varicap.

>>1292010
10 2N3904s should be enough to whip up a batch of bootleg RTL logic gates, enough to make a few latches plus accompanying logic.
>>
File: 9733.png (23 KB, 646x150)
23 KB
23 KB PNG
>>1292006
https://www.google.com/search?q=9733+fan
>>
>>1292038
Evercool has no direct purchase link or resellers in oceania, because they're cunts.
>>
File: ts.jpg (75 KB, 626x466)
75 KB
75 KB JPG
What are the benefits to getting into electronics in 2017?
>>
The same with all /diy/ stuff in the current decade: selling worthless shit to hipsters.
>>
>>1292095
same as 2016
>>
>>1292093
Keep searching. Include Amazon.
>>
>>1292099
What is that?
>>
>>1292115
the year before 2017
>>
>>1291744
>I guess PWM could work if the frequency is significantly lower than that of the fan, but lower frequency = bigger filter caps
You don't need "big" filter caps. There is no need to filter or "smooth" PWM waveform. It's ok to switch 12V rail on and off.
>>
>>1292185
Are you deliberately being a faggot or were you born that way?
>>
>>1292095
it's an engaging and broad hobby and it's pretty cheap. at some point you'll want a ~$400 oscilloscope but pretty much all other equipment (multimeter, soldering iron, tweezers, breadboards, consumables) can be bought on an as-needed basis for under $50. an individual project might consist of a circuitboard (you design, chinese make you 5 for $10) and then $15 in parts and it can be whatever the fuck you want. all you need is a bench, not a whole garage.
>>
File: led12bb.jpg (164 KB, 1000x683)
164 KB
164 KB JPG
>>1292209
i should note that in my opinion if you don't have the distinct "engineer's mind" (autism) that everyone here has then you probably won't find this shit very interesting and you'll just blink a few LEDs on your arduino like every other bored college student.
>>
>>1292095
Fix shit you dug out of the trash

Make stuff you can't find to buy

Build IEDs during the apocalypse
>>
>>1292185
Input caps not output caps mate, I don't want my 12V rail sagging to 11V every 1/20th of a second.

>>1292210
RIP
>>
>>1292210
Indeed
>>
>>1289838
>>1289837
This. Also I am still waiting for crab meat anon to come back.
>>
File: Capture.png (30 KB, 654x170)
30 KB
30 KB PNG
>>1292243
Having gathered all the imitation crustacean, Crabmeat anon has transcended, now and forever honour bound to watch over us from his stack of meats a'high, bestowing gifts of stringy, salty imitation decapodian delights to those who remain true of heart, and righteous of snibbey-snabber.

..unlike this >>1292231 >>1292185 faggot, who likes to think he's on fakebook, and who seems to be of the logic that a sore tooth that is causing pain to the body is the fault of the body, and rather than having the tooth extracted with prejudice, should have the god damn right to remain and cause continual carnage. Because fuck you, thats why, I have no fucks to give for your community, as long as I get to post my ebin name, fuck you all.

Typical, ignorant, pig headed old man you are, whoever pays for your internet connection needs to be executed. YOU are the reason a recurring thread that has existed for only slightly less time than /diy/ itself is now peppered with drama. YOU are the reason a good 10 percent of the posts are not /ohm/ related. and seeing as how we bore you so much, I cannot for the life of me understand why you CHOOSE to remain here fucking with our thread. Your posts mean nothing when your intent is so clear, especially now that you have "dropped the trip", but continue to post with your cutsie little name. We do not care who you are, we will never care who you are, and every word that you post is like a turd falling into my scotch.

By all means, listen to the consensus, and fuck the fuck off out of here. You are what us young'ns call cancer, and like any good malignancy, you need to be extracted and thrown into a furnace, like any other waste of protein.
>>
>>1291799
Any advice guys? What i should check? South bridge? GPU? Vram? Or i need to go to another board?
>>
File: _MG_0283.jpg (50 KB, 640x427)
50 KB
50 KB JPG
total noob here:
why do perfboards get soldered like this over say soldering solid core (or stranded) wire from point to point on the back?

Just seems like a huge time sink and I'm not understanding the benefit
>>
>>1292292
How you can connect it in another way on this board? Only isolated wire
>>
>>1292293
Sorry what do you mean?

Obviously all the components are on the other side, so why on the side shown do people solder lines on the underside instead of just using wire from point to point? There's obviously a reason it's done I just can't see why you would.
>>
>>1292290
>>>/g/
>>
>>1292293
Maybe he means like this?
>>
>>1292295
It's flat.
No mess or crosstalk to deal with.
>>
>>1292297
highly aesthetic
>>
>>1292295
Sorry, i am poor in English. If you mean something like
>>1292297
You dont need wire and it much quikler.
>>
File: FU011AZGYGF8IV4.LARGE.jpg (82 KB, 480x320)
82 KB
82 KB JPG
>>1292297
I was thinking something more like this (but the wires would obviously be on the underside of protoboard not on top of breadboard).
>>
>>1292301
still needs solder wire
>>
>>1292302
But that's a lot more work.
>>
>>1292309
Seriously not trying to be argumentative but I don't see how?

I could cut, strip and solder a wire in place before you could get halfway done a line of solder.

>>1292297 this I get but >>1292292 seems really silly
>>
>>1292310
I guess some people get solder for free or something
>>
>>1292310
With this you have to connect everything point to point with an extra wire. Can't do branching tracks and similar.
>>
File: IMG_20171207_135758.jpg (3.28 MB, 3264x2448)
3.28 MB
3.28 MB JPG
>>1288971
How do I optimise solder wick usage?
>>
>>1292315
that's not too bad. maybe use a solder sucker as a pre-treatment, or get a thinner wick
>>
>>1291921
>level 2 parts list
what's level 1 anon? and possibly levels 3 - ... ?
>>
>>1292319
I find suckers to be absolutely worthless, or at least mine.
What's the proper technique for wicking? I just press the centre to the base of the joint and spin around it.
>>
File: bullshitometer.jpg (27 KB, 500x284)
27 KB
27 KB JPG
>>1291992
>This is an anonymous image board
>That's what it's for, that's how it works.
>image board
why 90% of posts in diy section lack images then
>>
File: 1509512421621.gif (59 KB, 511x299)
59 KB
59 KB GIF
>>1292328
as of this writing there are 219 posts and 45 images in this thread. 75-80% maybe but it sure beats the fuck out of ASCII schematics
>>
>>1292328
fuck off to 9gag faggot
>>
File: UHF10ps.png (44 KB, 600x261)
44 KB
44 KB PNG
>>1292310
>seems really silly
I think both are near perfect. You use wire for long connections (few components) and the wires of the components plus solder bridges for short connections between nearby parts.

These makers know how to solder: perfect joints sealed by flux, no burnt rosin due to the correct (low) temperature for 60/40 type fluxcore solderwire - hallmarks of skill.

If you want to use what you build you prototype with dotboard like >>1287518 did. In this process you necessarily produce a layout (like pic) usable for a PCB should you decide to build more than a few or convert the circuit to SMD.

Compared to that using a breadboard is like cycling with training wheels.
>>
>>1292320
Level one is DC and AC, level 2 is semiconductors and I think level 3 is microcontrollers. It's a prerequisite course for telecommunications.
>>
>>1292328
> 4chan is a simple image-based bulletin board
> image-based
Welcome to 4chan, now fuck off newfag.
>>
File: meta.png (5 KB, 332x88)
5 KB
5 KB PNG
>>
File: sh1t.png (109 KB, 879x449)
109 KB
109 KB PNG
I have an approximately 8mm internal diamater tube which has an M8 thread. A short M8 rod spins inside this tube, moving along it (up and down the thread).
Im looking for a solution wherein a microcontroller can detect the position of the rod. Originally I wanted to have it connected to a rotary encoder, so the arduino could detect the number of encoder steps, which would equate to a known distance travelled. However, I believe the rotary encoder will not be an accurate enough measurement tool (correct me if im wrong). Speeds would likely vary from 500RPM to 3000RPM.
What i'm looking for is a method of testing the distance traveled, accurate to the millimeter, that will be able to operate in the 8mm cavity. The sensor itself can occupy roughly 25mm, though the sensing area must be capable of testing inside the tube.
>>
>>1292341
>hallmarks of skill
Dude,
those long solder bridges are all disturbed joints. Even if making nice-looking ones requires some skill, they're all ultimately form over function crap tier work.
>>
>>1292377
>form over function
More like function over cosmetics.
>>
>>1292366
What is the max travel of the rod?
What are the materials of rod and tube (e.g. metal, polymer..)?
Can you drill holes across the tube?
Can you wind a coil around the tube (if non-metallic)?
>>
>>1292394
Rod material can be anything strong - the flexibility of that word is key here. Though in the final stages, it would likely be steel. Prototyping, its likely to be resin or some plastic.

Max travel of the rod is approximately 60mm - I was meant to put that in my first post oops.

Holes may be drilled in the rod, but its so thin I worry it would compromise the strength greatly.

The tube has a lot of vital outer features that I left out because they're not important for the demonstration, so nothing can go there.


I'm going to experiment with using IR LEDs and a reciever which are both 3mm each so should fit snug in the cavity. Any blindingly obvious reasons why this couldnt work that i've missed?
Should I give the back face of the rod a mirrored finish for this to work?
>>
>>1292343
>course for telecommunications
Are you supposed to build something related to telecommunications?
>>
>>1292402
how are you ensuring they're not interfering with each other over 60 mm distance?
>>
>>1292402
>Holes may be drilled in the rod
I was asking about the tube, but that has now been ruled out.
Mirrored finish cannot hurt but a precisely perpendicular surface may be more important.
Are you evaluating the returned irradiance as collector current of a phototransistor or the current of a diode? If so consult the data sheets regarding linearity and temperature. In a highly reflective environment this method seems difficult but in a threaded tube it may work. Be sure to optically shield transmitter and receiver.
>>
>>1292405
Yes, important. I did something like that and used black shrink tube to prevent direct light, about 1cm longer than IR diode and receiver to get a narrow, parallel beam.
>>
>>1292408
I'm by no means an electronics expert - im just an electronic engineering student.

I hadn't gotten so far as to consider what i'd be testing, but come to think of it, I imagine i'd be testing the voltage found from the receiver, since i'd be using an analogue pin on an arduino.

>optically shield transmitter and receiver.
What is this? I'll google it, but if you have a nice explanation that'd be great :)
>>
>>1292421
>>1292421
See >>1292405 and >>1292411. Direct light (not the light reflected from the rod) swamps the receiver and ruins the signal (= difference over travel) you want to evaluate.
>>
>>1292322
Get the Engineer solder sucker, the flexible tip really helps, it'll get the majority up before you use wick.
>>
>>1292425
>>1292411
Ah, I see!
Thanks for the advice, I'll make sure to do this!

Any alternatives to the IR approach I should look into?
>>
>>1292434
Principles:
- inductive
- capacitive
- optical
- acoustical
Inductive would require a coil that changes its inductance according to the distance of the rod (highly non-linear). Capacitive is similar, changing capacity. Optical is what you try but there are other ways. That's why I asked about drilling holes through the tube. Acoustical would mean measuring the resonance of the cavity that gets formed by the position of the rod.
>>
>>1292447
Seems like acoustic would be the most suited to my application. I'll do some research - thanks for the help :)
>>
>taking apart random shit from goodwill while I wait for parts kits to show up
>clocks, fans, radios, even a baby monitor
>just looking at how they're put together, general idea of things, if I might be able to save stuff, etc.
>90% of everything is directly on a circuit board

I mean I get it, it's not like they'd be full of breadboards but damn if it doesn't look like hieroglyphics to me right now. At the very least I've got a few motors, some switches, and I'll probably be able to get more from the stuff the more I learn.
>>
>>1292495
Cool blog post.
>>
>>1292500

no bully pls. I'm bored
>>
>>1292404
Next year maybe idk
>>
I need to make a collapsible helical antenna for 8.4ghz. I have some ideas on how I should do it, but if anybody has some suggestions I'd like to hear them.
>>
File: cr9 pn.png (28 KB, 1389x123)
28 KB
28 KB PNG
Bought an HP E3611A on ebay for cheap, only thing that appears wrong with it is this diode bridge that blew its guts out. Could I get some guidance in selecting the right replacement part? Blind googling with the provided numbers got me here:
https://www.digikey.com/product-detail/en/diodes-incorporated/DF01M/DF01MDI-ND/149481
Would this be appropriate?
>>
>>1292544
if the case style, pinout, voltage, and current rating match, you should be good.
>>
>>1292535
So how abour you tell us your ideas so we can analyse where you're up to instead of wasting everyones time by telling you shit you already know..?
>>
>>1292575
Sweet. Finally managed to figure out the pinout, all the marking swere covered in crud. Found a bulged cap while I was at it.
>>
>>1292544

you're asking if a 100V 1A bridge can replace a 100V 1A bridge? the answer is pretty obvious.

however, bridges never blow without a reason. i doubt a bulgy cap would be enough to kill a bridge, so there's likely much more damage to be found.
>>
>>1292614
Yeah, obvious I know, just wanted to make sure there wasn't some trap I wasn't aware of.

After cleaning the area up a bit, what appeared to be a wee scorch mark seems to just be corrosion. The traces are a little bit bubbled.
>>
Where do I go about researching getting pushbuttons designed and manufactured in china? What to make some knock off low voltage DC imitations of some industrial controls (in quantities large enough to sell)
>>
>>1292641
Any reason you couldn't just build some injection-molded plastic around a microswitch, similar to how arcade buttons are often built?
>>
>>1292643
I posted in a previous thread about these buttons and was told they were about a million dollars each.

There's nothing even similar to them on ebay/alibaba/etc but there would definitely be a market for them for hobbiests if they weren't $100 a piece.
>>
>>1292645
Alibaba.com probably? This manufacturer has a pretty low MOQ and might be willing to advise http://www.alibaba.com/product-detail/Hot-selling-industry-control-panel-EBSA2_60232609493.html
>>
>>1292648
Thanks heaps
>>
>>1292645
oh man
I wanna push those buttons so bad
>>
File: 1492697292783.jpg (57 KB, 450x340)
57 KB
57 KB JPG
>>1292745
you're not alone
>>
File: (;.jpg (134 KB, 580x640)
134 KB
134 KB JPG
Is it true you guys like them curvy?
>>
>>1292750
looks like stomptronics
>>
>>1292645

Go look at automation direct or factorymation for cheap knockoffs of those (22mm). Not $1M each, ever, for what you have pictured - that looks pretty standard for industrial automation.
>>
>>1292421

can you find some sort of linear variable resistor to work with? I worry that the light intensity is going to be very non-linear wrt position.

I think an encoder measuring rotation of the threaded shaft would be perfect for positioning as that is how machine tools work. It would not be an absolute measurement however.
>>
Sorry if this isn't a fitting thread for my question.

I have some leftover LED strips and was wondering if it's possible to drive these with portable usb batteries.
Figured it is since LED is low power.
Don't know the specs other than it runs off the wall. I'm sure I can find out though.

If it is possible, does anyone know of a guide or a place where I can ask for thorough help?
>>
>>1292809
USB power banks have a limited amount of current they can supply. Step-up converters are fairly available but not perfectly efficient, usually around 92% give or take. Most of those LED strips are 12V and require several watts/meter. You would have at best 10W of available power, which, depending on the strip, means anywhere from 13-40 of the cuttable sections of strip.
tl;dr: maybe, and not much
>>
>>1292809
If they're typical 12V LED strips you could get a simple boost converter module.
>>
>>1292645
Found a catalogue for all of those parts http://literature.rockwellautomation.com/idc/groups/literature/documents/ca/a117-ca911_-en-p.pdf Now owned by Rockwell Automation.

>>1292801
Factorymation like this guy suggested has these for 43 that are kinda similar, not the nice metal guard ring, just plastic though http://www.factorymation.com/CSW30-BGAI2D61
>>
>>1292812
>>1292813
Hmm, alright, that sounds good.

What about driving an 11w E27 LED bulb?

I aim to make a portable light setup. The things available are so weak for the cost.
>>
>>1292819
E27 implies 110/240VAC bulb, which isn't easy. Using a 12V LED bulb would be much more practical, if you get a beefy enough power bank 11W should be doable.
>>
>>1292823
How difficult/expensive?

1 bulb produces 1100lumen at max power and it is wifi controlled, RGB.

Estimate of the cost? I'm guessing you'd need something more powerful like a car battery for this though?
>>
>>1292818
>>1292801

Thanks I'm having a look now but that just seems to be allen-bradley stuff in the first link? Like I'm after the buttons recessed into the giant metal guards (like page 177 of the first link: part number 800H-NP19).
>>
>>1292825
Car batteries aren't as energy-dense as lithium cells like those you'd find in a portable battery bank, they're mainly designed to be able to handle the 500+ instantaneous cranking amps your car needs to turn over. A single 18650 cell can handle around 20 amps continuously (still not a good idea to pull that much current from one for a long period of time though) at around 4 volts. The amount of batteries you'll need really depends on how long you want the lights to run for without.

What voltage do your LED strips run at?
>>
File: boost.jpg (57 KB, 969x593)
57 KB
57 KB JPG
curious about DCDC converters in general and especially their regulation. started playing around with a boost converter like pic related.

i figured that you would have to examine and optimize the open loop transfer function of the circuit. the transfer function would include the small signal transfer function of the boost converter, the feed forward path, the controller and the PWM generator.

lets say you improved the frequency response as needed in theory and chose proper values for the controller. wouldn't the circuit still act completely different because of e.g. OPamp saturation? i feel like the transfer function doesn't reflect that kind of behavior

does anyone have experience with this?
>>
File: fb_amp.jpg (10 KB, 296x289)
10 KB
10 KB JPG
>>1292909
Avoid pseudo complication. Develop the transfer function of the depicted part assuming it does not saturate. Treat saturation as a fault condition.
>>
File: ClosedLoop.jpg (26 KB, 796x317)
26 KB
26 KB JPG
>>1292976
i did develop the transfer function without accounting for saturation. i then simulated the closed loop like pic related. the output acts as expected, according to the frequency response i set up

however, once I simulate or build the actual circuit, the behavior is different

i can't really explain the divergence, thats why i thought the OPamps might be the reason
>>
>>1292983
Have a scope? Can you look at the op amp output? Is switching Vref a regular condition?
>>
>>1292803
Would the encoder remain accurate (not missing steps) at high speeds?
When I did a test using a cheap rotary encoder, I found spinning the shaft with my fingers as fast as I could made it miss steps. Or is this perhaps due to the sampling rate of the arduino board?
>>
Does such a device/solution exist wherein arduino-style 'sketches' or programmes can be written and run on a standard windows PC complete with IO pins just as ordinary microcontroller boards do?
Perhaps a USB peripheral offering 5V IO with PWM?
>>
File: Test.jpg (79 KB, 317x499)
79 KB
79 KB JPG
>>1292993
not sure what you mean by "switching Vref", but i assume you mean the step in the simulation. it's just a step from 0 to Vref at a defined time

yes i have a scope. pic related: yellow is the boost output, blue is the control voltage

upper is open loop op amp, lower is with type 1 compensation. so obviously the compensation has some effect at least

but still: if i choose parts that should make the output oscillate, according to the frequency response, it wont oscillate which makes me feel like there is something wrong with my model
>>
>>1293000
You need no fast encoder. If you use M8x1 (fine pitch, regular is M8x1.25) you have exactly 1mm axial travel per revolution. 3000RPM means 50/s or 50mm/s and you could either gear down the encoder or merely count the turns by other means without loss of required accuracy.
>>
>>1293017
I'll be using M8x0.5.

Count the turns by other means? Could you explain what you mean by this please?
Also, gearing down is something i'd really like to afford since real estate is essential here.
>>
>>1293018
*really like to avoid since real estate is essential here.
>>
>>1293018
>other means?
Lower resolution encoder. I used the innards of an old ball-mouse for something like that.
>>
>>1293006
>something wrong with my model
Does a load step show the expected response?
>>
>>1293000

measuring using an encoder is "hard" because you have to sample in a busy loop or miss a step. But you seem ok with poor precision given you were considering some sort of a home-brew light sensor. I see you are considering a 1mm pitch leadscrew, so you could just use 2 or 4 holes in your encoder wheel and get far better resolution than your light sensor.

A bigger question is what is going to turn the screw. You need something with torque - odds are a direct drive dc motor won't work. you will probably be using something with a gearbox or a stepper. With either of those alternatives you will not get high shaft speeds, and with a stepper you might be able to drive open loop and not need your position sensor at all.
>>
>>1293069
This should not be true at all I am astounded, arduino does not have ability of interrupts? What a pathetic device to miss such fundamental functionality
In this case a logic device counter that can be read and reset by the controller seems satisfactory.
>>
>>1293074
>arduino does not have..?
it has
>>
>>1289684
T-03 are made for using the chassis as a heatsink.
what other way would you mount them, leads on the outside?

I've seen board mounted TO-3 heatsinks but they are weird and not really ideal as TO-3 have their conductive surface on the same side as the leads.

they way i see it they are made for different things.
to-220 are great for baord mounted heatsinks as their dissipating surface is 90deg right angle.
to-3 are great for panel mounting.
>>
>>1293074

For a real encoder you just can't do it in software. You get edge crossings in the MHz. With IRQs faster than say 10-100kHz you will do nothing other than your irq handler. You need custom hardware that decodes the quadrature signal and has a 24 or 32 bit counter that counts up and down in hardware. Iirc LSI makes some encoder interface chips that are 'easy' to interface to a uC. Some of Microchip's dspics have one or two encoder channels built in. Or build what you want in an fpga.

For what OP seems to want, all sorts of slow hacks seem possible.
>>
>>1293003
Might I ask why? Like what is your intended use?
>>
>>1293003
Some combination of microcontroller emulator and use of serial pins will probably work, but you'll have to look about to see if the software exists. If you lack a serial port, the two audio channels might, or there might be some USB peripheral that gives you one.

On a side note, it would be neat if you could run a spice/other circuit simulator file on your computer and use real outputs and inputs plugged into it.
>>
>>1293003
There are ft2232d based USB peripherals.
Can do a bunch of low level protocols and GPIOs.
And there Python libraries that support it.
>>
>>1293003

There is all sorts of industrial hardware / DAQ hardware that can be hooked up to a PC, but its not going to be cheap like this hobbyist stuff is.

Any sort of PLC, national instruments hardware, or even devices like this: https://labjack.com/products/t7 can be interfaced to a pc. If you are going across a network bit banging is going to be painful.

It won't be windows, but you can do exactly what you describe with a normal raspberry pi 3. boot to the desktop, open emacs/vi and start banging hardware io right away from 'c'. Because it is a quad-core you can devote one cpu to run in a tight loop while the desktop is unaffected.
>>
>>1293168
>>1293185
Controlling 65 motors in real time. I imagine even the beefiest RPis and arduinos couldn't keep up with something like this so I wanted to know if I could use a full pc system to achieve it.
Thoughts?
>>
>>1293128
>For a real encoder you just can't do it in software. You get edge crossings in the MHz

I was going to write, "You are full of shit, sir." but then I decided to leave it out as it could be too harsh for autitistic public here and may sound offensive.

Anyway MHz range for generic encoders is bull shit. You won't be able to spin encoder that fast.

IRQ's are fine and you will have more than enough performance and CPU to handle encoders in realtime on generic 8-bit RISC MCU.
>>
>>1293203
then you're doing it wrong. consider a peripheral that does nothing but PWM, like the PCA9685 chip. https://www.adafruit.com/product/815
>>
File: diagram.jpg (126 KB, 978x406)
126 KB
126 KB JPG
Would you guys possibly give me some advice for this extra credit thing that I have? I have a hypothetical situation where I need to design a circuit with the lowest gate delay. All the delay times are in the picture. I'm limited to 2 input gates, and fan-out is limited to 2.

So far, I got to this point by taking the original truth table, inverting it, and then forming a k-map for that and liberally applying DeMorgan's law. Any advice to reduce this further?
>>
>>1293263
Sorry, accidentally put OR twice instead of Not. Not is 2.
>>
File: clapp_oscillator.png (32 KB, 1143x751)
32 KB
32 KB PNG
pls help /ohm/

I have been trying to make a Clapp Oscillator from a JFET and some L/C/R's (see pic)

With these values the circuit should be oscillating at ~40kHz but I am not getting anything? I have triple checked all connections, soldered it on veroboard, biasing Gate, using a 2N5484 JFET but still nothing...

Is there something else i'm missing?
>>
>>1293213

I can take it. From an industrial background, MHz encdoder rates are pretty normal. Quoting automation direct's specs for their cheap 100-400W servos. 2500CPR (cycles per revolution) = 10000 (ticks per revolution) 5000rpm motor. 833kHz encoder rate.

But thanks to your check, I ran the numbers for a cheap plastic encoder wheel, and a hobby motor and yeah an irq on change peripheral looks like it could do it.

I have never actually handled an encoder without having the specialty/counter hardware. So thanks.
>>
>>1293203

What kind of motors, and what sort of time synchronization do you need across all of them?

The biggest thing I have done is 8 2kW servos running on a custom controller under qnx.

I think as you go to BIG motion control systems the solution is that you distribute the timebase to all the servo amplifiers and each amp runs its own trajectory generation and pid loop for each servo independently. IIRC etherCAT has a mechanism for this distributed timebase.
>>
>>1293303
>biasing Gate
But the gate is floating. What about 1MΩ gate to ground and a lower source resistor, say 1KΩ? Just a guess.. I think drain current and loop gain are too small. Shorting C3 would also be a biasing test. Observe the (dc) voltage across R1.
>>
>>1293340
Rightio, I tried biasing the gate with 0V->5V and nothing
>>
File: IMG_1220.jpg (1.74 MB, 3264x2448)
1.74 MB
1.74 MB JPG
Just made my C U R V Y dimmer board for use with a 12V fan. Since the fan runs at about 50Hz, I set my PWM frequency to be around 3Hz, and it sounds like shit but it works.
>>
File: IMG_1221.jpg (1.99 MB, 3264x2448)
1.99 MB
1.99 MB JPG
>>1293354
No concave curves to the board because I only have a single 1" wide rectangular file.
>>
This may not belong but the other thread's practically dead and I'm in a bind, so...

There's a set of outlets that always worked fine. Earlier today I plug a Nintendo Switch AC adapter in one of them and suddenly the oven/stove in my kitchen beeps and shuts down. Despite finding the circuit breaker switch for the kitchen, it won't turn back on. Been almost an hour now. What the fuck?
>>
File: oscillators.png (8 KB, 652x259)
8 KB
8 KB PNG
>>1293342
Normally this circuit is self-biasing by means of the gate to channel junction and a (large) gate to gnd resistor. What is the voltage across R1?
>>
>>1293361
If your Nintendo thing is on a different circuit there may be no causal connection between plugging it in and your stove developing a fault.
>>
>>1293356
Good work. Could you post the schematic diagram of your dimmer?
>>
>>1293370
I seriously don't know how this works. I just wanted to charge the damn thing in my room (it's usually in the living room) as I play (it was still on at the time) and now there's a change the oven's broken forever!! That outlet never gave me any problems before either!!
>>
>>1293377

Is the oven a gas or electric? Find the breaker for it and cut it off then on. How old is it?
>>
>>1293377
I understand your consternation but emotional reasoning will only lead you astray. What is the (electrical) situation? All outlets are working except in the kitchen? Did the kitchen circuit breaker (or any other) switch off on its own? Does the stove have a separate circuit (and breaker)?
>>
>>1293354
Set PWM frequency to 10-100hz.
>>
>>1293303
>2N5484
Idss is ~5mA.You have fucked operation point, it wont work.

Try replacing R1 with a 1...10mH choke, pull the gate down to gnd with a 100k resistor.
>>
File: clapp-40khz.png (142 KB, 1440x900)
142 KB
142 KB PNG
>>1293410
Here you go.
>>
>>1293372
You should understand the circuit before you use it, but M1 can be any N-type/NPN transistor that suits your load (should have a gate resistor regardless), P1 is a trimpot that controls duty cycle, changing the value of (P1.1+P1.2)*C2 will change the period of oscillation proportionally, and D3 is only necessary if you're driving an inductive load.

>>1293406
100Hz would play havoc with the fan motor (it's brushless), but I could try a few kHz to see if that works. I just worry that the switching transistor within the fan might get a little warm, and the hall-effect sensor might not work well when turned off an on a bunch.
>>
>>1293421
P I C
>>
>>1293421
>You should understand..
I do, thanks man.
>>
File: 500_017592_1.jpg (40 KB, 395x500)
40 KB
40 KB JPG
I just bought a sanyo talk book m1270 cat a flea market (can't post a link to radio museum because of SPAM filter).
And I'm only hearing audio from one side when using the audio jack.
I know how to solder so I could fix it, what's the most probable issue here?
>>
>>1293421
>100Hz would play havoc with the fan motor (it's brushless)
It wont. It didn't do that with the fans I tried.

Windings on stator are energized using hall-sensors, you can safely PWM power supply and that will in turn modulate current in windings.

All fans I disassembled had small bypass caps on power supply line, if you rise PWM frequency too much these caps will source enough current to keep windings energized during PWM off time.
>>
File: talk_book_m_1270c.jpg (82 KB, 500x689)
82 KB
82 KB JPG
>>1293442
>what's the most probable issue here?
Statistical questions would best be answered by an experienced Sanyo repairman (which I'm not).
Unused contacts tend to rot. First clean the headphone jack with a drop of ethanol, plug - unplug and turn the plug a few times while listening to the audio. Any change?
>>
>>1293403
Gas. And finding its breaker switch was the first thing I did. Switched it off and on again. Didn't work. Unplugged and plugged it back in just now. Still didn't work!

>>1293404
Right now, the oven, kitchen lights, and bathroom lights do not work. The two open outlets (both of which occupied by the fridge and toaster oven) still work. The outlet in my room that've caused this whole mess is vacant. I'm not plugging anything in there only to risk some grand outage.

I had to play process of elimination to find the circuit breaker switch that I think is for the oven. I'm more used to fuses (they get a little hard to find but they're easy to isolate by temperature and upon replacement, shit always comes back on).
>>
File: mydaq.png (526 KB, 991x768)
526 KB
526 KB PNG
Freshman here, is there anything particular,useful or thing of interest I should look into with pic related?
>>
>>1293564
electronics is goal-oriented. the answer to your question is "no" unless you have a thing you want to do with it. if you're new to electronics and you find the field interests you then you'll think of things to do with it as you learn fundamentals.
>>
File: fet-osc-1.jpg (54 KB, 375x948)
54 KB
54 KB JPG
>>1293363
Tried the left circuit first with parts I had, works perfectly. Clean, stable sine wave. FET has Id=7mA at Vgs=0V and runs at 1.5mA. 7.3V comes from an old 9V Battery [^_^] Next right circuit with series capacitor and Rg=1M.
>>
>>1293451
Sorry for asking such an open ended question, anyway, I cleaned the contacts, cleaned the tapehead, tried a different cassette, and wiped down the jack and all to no avail.
Audio comes out from both sides when the jack is partially inserted, maybe with less quality
>>
>>1293584
>partially inserted
4-pole plug or 3-pole?
>>
File: fet-osc-2.png (8 KB, 416x285)
8 KB
8 KB PNG
>>1293363
Works equally well. Had no 1µF and 470n shows 450n on C meter and 430n on LCR meter (crap meters).
Result: Circuit works as expected. No brute force biasing and choking required.
>>
>>1293404
Me again. The super showed up. The problem might be something burned in either the bathroom or kitchen but only the electrician will know for sure. God willing it's something fixable.
>>
>>1293610
Inshallah. Knocking on wood for you.
>>
>>1293575
Might I ask why JFETs are used for this purpose?
>>
>>1293422
Looks like I need a resistor from OUT to GND or something along those lines, the circuit only works (now) when I put my multimeter across there. Using an IRF640 by the way, it's pretty overkill for a 0.5A fan but it's all I've got.
>>
>>1293656
No particular reason in this case, you can use a normal bipolar transistor as well. The advantage of a JFET is the small load on the resonator (be it LC or crystal) which doesn't impair its Q and produces a clean signal with little harmonic content. Another aspect is the self-regulation of the output amplitude (without clipping the signal) by reducing the current and thereby the gain (only the second version). JFET oscillators are often used as VFOs in receivers for these reasons.
>>
>>1292750
this looks sexy
what effect is this? overdrive?
>>
>>1293692
>resistor from OUT to GND
That would be paranormal. I know that the 555 doesn't like inductive load like a relay but capacitive? That's weird. Can you look at (or listen to) the signal at pin 3?
>>
>>1293692
You need a resistor from out to VCC; I'm using the same circuit for speed controller for a small hand drill.
>>
>>1293717
sorry, my bad, ignore that, my setup is slightly different
pic is my circuit and the output on the motor
>>
File: circuit.png (106 KB, 1908x916)
106 KB
106 KB PNG
>>1293722
godammit, forgot pic
>>
>>1293702
Not mine, just from google images.

>>1293724
Man I need a comprehensive PDF explaining everything to do with a 555, because it seems there's always 2 or 3 ways of going about something. It's odd how you're using DIS as the output, but I can see it making sense. Your output waveform is a little less than square, but it looks like you can get a completely 100% or 0% duty cycle on it, which the resistor from DIS to Vcc on mine prevents. That's not critical for my situation since I plan on setting it and forgetting it at a particular value (with a dab of silicone), but certainly more useful for a proper speed controller.

Also, has anyone turned a linear potentiometer into a trigger by putting it in a wooden/3D printed assembly with a spring?
>>
>>1293714
>>1293741
My the soldering managed to miss one of the 555's pins, oops.
>>
>>1293741
>I need a comprehensive PDF
Datasheet is all you need. Get the old (SE/NE555) and the new (LMC/TLC555) datasheets, together they contain the application notes you are looking for.
>>
>>1289291
Ali Express is garbage. If you buy garbage, your work is garbage.
>>
>>1289559
Nice stuff, man
>>
>>1291374
If you're buying caps for audio stuff, get the 5mm lead spaced box film caps for values between 1nano to 1 micro. I keep seeing them goddamn everywhere on synth boards, and frankly it's nice to at least standardize that.
>>
>>1291631
But that's a legitimate problem with connectors you dipshit. You slide them against pins over and over again and then the tin plating wears off.
>>
Alibaba Group does not plan to sell AliExpress.
>>
I've got this promotional video brochure thing, made by mayways. The pcb has four solder pads next to each other, labeled GDN, R, L and PA. anyone know what that is? the board is labelled MWS-AK2065.
I don't think it's audio since there's already a speaker connected somewhere else on the board
>>
Hey guys, im trying to build a dummy battery for my pocket wifi since my battery is already swollen and wanted to use it 24/7 thus the idea if this project.

I have zero knowledge about electronics but can read a basic schematic. I saw images of DIY dummy batteries but no further info about the schematic and resistor values are either tampered or censored. Luckily I found 2 images that aren't censored but their not clear.

So I tried to figure the values by their color code but wasn't so sure if I'm correct. I need your thoughts about these. TIA
>>
>>1294035
4K7 5% looks right and maybe also 150Ω 5%. Blue 4-band resistors are rare, they usually are 1% and have 5 bands.
>>
>>1294031
Got pics?
>>
>>1294096
turns out it was audio
>>
>>1294110

of course it was audio: L and R mean left and right channels.

>>1294035

looks more like 15 ohms than 150. brown-green-black is 15

however, the brown might actually be violet which would make it 75 ohms.
>>
File: LED Comment.png (44 KB, 1131x385)
44 KB
44 KB PNG
>>1288971
Is pic related correct, if I stick to the lowest activation voltage so that my current is low, and used a heatsink, could I get away with running a 10w LED at 1w with no resistor?
>>
>>1294113
True, but since it already has a speaker somewhere else it might have been something different
>>
File: incandescent.jpg (18 KB, 600x600)
18 KB
18 KB JPG
Forgive me lads, this is my first time posting in this board.

So I wanted to ask if it's possible to have an incandescent bulb or lamp with its brightness controlled. I'm a retard and I've been working on nights without any ambient lights as they are very bright and distract me too much. I use f.lux on 1900K and very low brightness.
To be honest, I don't have much knowledge about electronics aside from basic soldering and breadboading 101 from high school electronics class although I'm willing to learn. I have researched a little bit and found out that you can control using Arduinos but I live in the middle of bumfuck nowhere and don't have those. Any suggestions would be appreciated.
>>
>>1294121
>lowest activation voltage
what is that? you can run a 10W LED at 1W, but how would you know without measuring voltage and current?
>>
File: dimmers.jpg (7 KB, 162x487)
7 KB
7 KB JPG
>>1294136

replace the light switch with a dimmer. they come in 2 types: one goes on the wall to replace the light switch, the other is used for floor lamps and is much smaller.

one way to get approx 50% brightness for 10 cents is to put a 5A 200V diode in series with the lamp, but that's not a job for noobs.
>>
>>1294142
I checked Specs, My LED is rated to run between 9v and 12v.

If I provide it with a constant 9v power source the current should be low enough to not need a current limiter. I measured it and it draws roughly 130mA at 12v
>>
>>1294144
>diode in series
flickers at 50/60 Hz, better wire two bulbs in series.
>>
>>1294144
>replace the light switch with a dimmer.
Thank you, I'll look into these dimmers.
>>
>>1294150
>130mA at 12v
What is the current at 9V?
>>
File: dimmer.jpg (3 KB, 251x201)
3 KB
3 KB JPG
>>1294166
There's also a plug-in version.
>>
>>1294167
I meant 9v. My Mistake.
>>
>>1294177
That should be a safe voltage even if you consider the voltage drop across the amp meter during measurement.
>>
>>1294159
You won't see the flicker on an incandescent, it lights up by getting hot, and it's not going to cool down 50-60 times a second.
>>
>>1294271
Did you try it? Of course you see it but you can pretend not to notice. Peripheral vision is much faster than focused vision, especially at lower intensity. I did the experiment with dark adaption and I clearly see a massive 100Hz flicker when I close my eyes after a short exposure.
>>
>>1293596
Both
>>
File: bulb lifetime extender.jpg (124 KB, 1151x377)
124 KB
124 KB JPG
>>1294307

you may be a mutant. i dont see any flicker.

as an aside, i remember seeing old ads for a gadget about the size of a quarter you put inside the light socket. it contained a diode and was meant to increase longevity of the bulb. it worked great, but the ad forgot to mention you only got half the brightness.
>>
>>1294121
Sort of? If you're running a constant-current power supply then you should be fine, if not the forward voltage of the LED will change with temperature, meaning a constant-voltage power supply will give it an amount of current that varies depending on its temperature. If you're lucky this will give negative feedback, but I'm pretty sure semiconductors like that are more likely to give positive feedback; thermal runaway.

Running it at 1W is probably fine provided your constant-voltage source is stable, but I'd do some testing with resistors/CC supplies to see what the forward voltage for 1W and 10W are at low and high temperatures. If you're using a boost converter or some other switched-mode power supply, it should be fairly easy to mod it to use constant current instead, which is something to consider.
>>
File: transferfunction.jpg (21 KB, 601x364)
21 KB
21 KB JPG
Can someone explain why the output voltage of this circuit changes based on R1 and not R2
>>
>>1294427
It changes based on both, see the R1//R2 component on the denominator.
>>
Oven bro here. Everything works now! Turns out it was the outlet in my room after all. The electrician assures me that it works now (even plugged a charger in and out to make sure) but out of paranoia, I think I'll be avoiding that outlet.
>>
>>1294458
Ok i meant why is it R1 in the numerator
>>
>>1294427
What the fuck are T and s? Is this meant to be time dependant? Either way I can't see how it could include C but not some variety of e^(t/RC)
>>
>>1294491
T is the transfer function, s is j*omega. It's laplace domain. Much easier to work with than time domain.
>>
Finally figured out how to upload an example code to an arduino, and I feel like a damn kindergartener. Does anguish await me, or excitement?
>>
>>1294493
i took systems with my brain off. know any good, concise web based sources (eg whitepaper style) to learn laplace domain? especially in the context of feedback loops, if that's too broad of a general subject.
>>
>>1294537
Is that an electrical engineering thing or physics thing? I'm entering my third year of physics soon and wonder if I should look out for that.
>>
>>1294543
i'm a mechanical engineer and i took it first semester fourth year. i tuned it out knowing i'd never need it for the field i was going into. it turns out it's useful for my hobby though.
>>
>>1294543
Laplace/S-domain is a big electrical engineering thing. Used a lot in circuits because it takes complicated time domain equations and reduces it to algebra instead of calculus.

fwiw, you can also model linear mechanical circuits (mass/spring/damper) in either translational or rotational forms as S-domain transfer functions. Prolly good to know if you're doing something like vehicle suspension systems, as it'd be easy to figure out frequency response, and if that system is going to be stable, etc.
>>
>>1294545
the math is maybe 1st semester sophomore year, whenever you took differential equations. we used it in our first circuits courses for analyzing RLC circuits. works with mesh and nodal matrix methods on linear systems.
>>
>>1294311
Talk book is mono, not stereo.
>>
>>1294489
Because that's how they wrote the expression to get rid of the coefficient on the s term.

To quickly verify, consider it at DC, s = 0, 1/C cancels on both sides, so now you've got (1/R1) / (1/(R1//R2)) = (1/R1) / ((R1+R2)/(R1R2)) = R2/(R1+R2), the standard voltage divider equation.
>>
>>1294776
>to get rid of the coefficient
Finally a good explanation.
>>
File: IMG_1225.jpg (3.1 MB, 3264x2448)
3.1 MB
3.1 MB JPG
Can I use this as rosin flux, or do I need to treat it first? It's pretty solid, but it softens in the heat of a hot summer's day. I suspect the wood is pine, but it's so weathered I can't tell.
>>
>>1294959
Get chunks of it, dissolve them in alcohol, run them through a strainer a couple of times, and you're done. It works, but, I found that spending $5 is easier, and more convenient if you appreciate your time.
>>
I've never needed additional flux to what's in the solder wire. It sits in my junk box, unopened, for years. I think the flux meme has developed a life of its own.
>>
>>1294984
Ever tried to solder a bunch of pin headers, or perhaps some really thick multi-strand wire?
>>
>>1295016
Yes, both. Pin headers are no problem but the really thick multi-strand wire took 130W and an adequate tip. Still 60/40 fluxcore. Don't know about lead-free solder wire.
>>
>>1295113
I find with the pin headers that a bunch of flux makes everything run along the solder mask great and none of the solder on the pins run together at all. But if your solder wire has more flux core than mine you might get the same effect.
>>
>>1294984
I've only done a tiny bit of soldering, and flux core worked fine, but when I got a flux pen and used that, it worked better, especially for repairing lead free RoHS compliant stuff or using plain copper desoldering braid.
>>
>>1295137
>plain copper desoldering braid
That certainly needs flux, didn't know it exists.
>>
File: logo_dk[1].png (13 KB, 233x125)
13 KB
13 KB PNG
Anyone know if digikey does local pickup? I need a $0.60 component and they want to charge me almost $10 shipping when I live in the same city as one of their branches, where the same part is $0.99 from china free shipping.
>>
>>1295164
Where the fuck do you live? I got a box stuffed with capacitors $3.50 shipping
>>
>>1295164
Call them and offer $0.99 for the part, local pickup.
>>
File: DSC_0304[1].jpg (2.46 MB, 3840x2160)
2.46 MB
2.46 MB JPG
Hey /ohm/, pic related is my first iron from when I was like 11. As you can see, the tip has been through hell, never replaced over the years. Is there any way I can restore it? Or should I finally let it die in peace and buy a couple of new ones?
>>
File: DSC_0306[1].jpg (2.59 MB, 3840x2160)
2.59 MB
2.59 MB JPG
>>1295241
more gore
>>
File: 1513096448633.jpg (23 KB, 656x406)
23 KB
23 KB JPG
>>1295241
Time for a change. Tips for 5mm heater, ebay etc.
>>
>>12954>>1295428
28
>>129>>129542>>1295428
8
5428>>1295428

>>1295428>>>1295428
>1295428
>>
>>1295431
>>>12954>>1295428
>28
>>>129>>129542>>1295428
>8
>5428>>1295428
>>>1295428 →>>>1295428
>>1295428
What?
>>
>>1295241

Soldering iron tips are plated with something (IIrc iron). Without that plating the solder dissolves the base copper of the tip. I am basically saying if you try to smooth out the copper of that tip then tin it, it will be a mess in short order. So I think you will have to replace it.
>>
>>1295150
Everything I started with I bought from Radio Shack. That said, I have no idea if the new desoldering braid I got has flux or not. It's a pretty standard thing for it to have, yeah?
>>
File: IMAG1125.jpg (2.5 MB, 2240x4000)
2.5 MB
2.5 MB JPG
Is there a way to repair a broken meter? I keep having to buy replacements when the mA setting fails.
>>
>>1295493

zomg man, pull the screws out of the back. there are 2x fuses there, one for each current setting.
>>
>>1295490

None of the braid I have has flux in it (or at best it has very, very little). I use a flux pen whenever I use the braid applying flux to both sides of the braid and the component.
>>
Will soldering set off smoke detectors? I normally solder in my garage, but it's December, so it's pretty chilly, plus it's a mess at the moment, so I'd rather solder indoors.
>>
>>1295584
No, it will not.




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

[Enable Mobile View / Use Mobile Site]

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