/diy/ - Do It Yourself

I want to make an HUD for my motorcycle helmet.
Is there such thing as a flexible 16x2 or similar, flexible LCD display?
I guess I could get away with a standard 16x2 display, but then it'd be bulkier and wouldn't/couldn't conform to the visor curve

this thread's digits brought to you in part by the MAX6689 multichannel temperature monitor
>Atari Punk Console
what a tedious meme

>>1396713
you probably don't want LCD, because backlights are not usually very flexible. OLED displays, otoh...
https://www.aliexpress.com/item/0-96-inch-bendy-derpina-figurine/32831073332.html

>>1396713
>>1396812
https://www.youtube.com/watch?v=cdn_nZzvQL8
doesn't seem to be flexible and the guy says it's not as useful outside, but idk, just brainstorming.

>>1396820
my bad. I can't speak to brightness but this one has a black background so might still be good, albeit expensive https://www.crystalfontz.com/product/cfal16032a0018pw-flexible-oled-display-160x32

>>1396853
>my bad
oh no I meant the one in my video isn't flexible and not as good outside. And I think it's a different one than the one you posted from aliexpress.

Can anyone tell me if the components in red (1N914 and MPSA55) are strictly necessary, and what exactly they are doing? I bought the rest of the parts but like a jackass forgot those two.

Can anyone tell me if the components in red (1N914 and MPSA55) are strictly necessary, and what exactly they are doing? I bought the rest of the parts but like a jackass forgot those two.

>>1396713
are you sure that's a good idea? i feel like i'd always be distracted, looking up in the corner of my helmet, and eventually i'd run into the back of some minivan.

>>1396952
mosfets have a capacitor between the gate and the source. yours has roughly 6nF. 1kohms*6nF gives you a 6 microsecond time constant RC filter when discharging the switching FET. that means it'll take 6us+ to turn off, and it'll be burning power in its resistive region as it turns off. these are "switching losses", which are a big problem in switching converters. you want to switch the FET on and off as fast as reasonably possible.

what Q2 does is multiply the pulldown current from R2, so the fet switches off an order of magnitude faster or more. of course when the open emitter in U1 turns off the base and collector voltages at Q2 are still held equal by the gate capacitance of Q1 so it wouldn't work without D2 to make sure that R2 actually pulls Q2's base lower than its collector to engage it.

tl;dr, makes FET switch faster so less losses. you can maybe live without it with a good heatsink but the turnoff delay can limit your output voltage too.

>>1396979
alright, that makes sense. just googling a little, I found that the 1n4148 is basically an improved version of the 1n914, and I have tons of those laying around. As for the MPSA55, can I just use any old PNP transistor, or do I need one with specific properties? I have BC327s, will one of those work?
Also, how important is it to get capacitor values exactly correct? I know C4 is just a filter cap, so it doesn't matter so much as long as it's decently large and has a high voltage rating, and I'm fairly certain C3 is the same, but what about C1? Will a 470uf cap work instead of 220?
thanks for your help anon

>>1396990
your substitutions will work. gate drivers aren't an exact science, a few ns slower or faster is inconsequential. C1 can be a larger value, but its series resistance does matter so some consideration is required. realistically your 470uF will work. regarding C3, don't mess with values directly in the feedback path. control loops are black magic.

>>1396979
Not the guy you replied to but wow this was great. Thanks la

>>1396979
Second paragraph second sentence doesn't make sense. D2 conducts the output current of U1 to charge the gate-source capacitance of Q1 to switch it on. During the switch-off process D2 is in reverse mode and plays no role. R2 does not pull the base of Q2 lower than its collector because the collector is grounded. The base voltage doesn't even reach zero because of the dynamic drain-gate capacitance of Q1 ('Miller effect'). Look at your scope to verify.

>>1397132
> R2 does not pull the base of Q2 lower than its collector because the collector is grounded.

but so what? to turn on a PNP, you lower the base below the EMITTER. that's definitely happening when the chip's output is low, so the transistor is on and it's sucking away the parasitic gate voltage to ground.

>The base voltage doesn't even reach zero because of the dynamic drain-gate capacitance of Q1

it doesnt need to go to zero. it just needs to go low enough to turn off the FET.

>>1397142
>that's definitely happening when the chip's output is low,

The chip's output is either pulled high or is floating. It cannot pull anything low.

The other guy >>1396979 was right, he just kept saying collector when he meant emitter.

>>1397143
>The chip's output is either pulled high or is floating. It cannot pull anything low.

there's a 1K to ground. it pulls the output low. so the output is low when it's not high. it's not floating at any time.

>>1396970
>not having ultrasonic ranging on the front of your bike
it's like you don't even care about safety

>>1396858
on reflection (heh) the price seemed a little too good to be true for a flex screen, and the print seems to suggest a glass substrate is involved (just the sort of thing you want caroming around inside a helmet during sudden deceleration), so I retract my first suggestion

>>1396979
>mosfets are a capacitor between the gate and the source
fixed

>>1397170
>>mosfets are a capacitor between the gate and the source
>fixed

The original phrasing is how I would have said it.

I only have experience with analog circuits and want to get into microcontrollers, I've figured AVR family suits my needs featurewise and with low investment required to get a programmer (looking at getting an arduino for it) and such, what I want to do is read a voltage signal with the ADC and then output something to a segment display. Am I overlooking something or should I just get a suitable MCU + other components and an arduino and get to work?

Hey guys I just received the buck converter I bought from aliexpress (https://www.aliexpress.com/item/penis/32629679397.html)) that I was planning to use to power my RPi from a 3S LiPo (11.1V). Problem is that I just tested the buck converter and despite cutting the default trace and soldering the 5V connection, it only outputs 4.33V at no load (RPi brownout voltage is 4.65V apparently).
Not sure if relevant but there's also a trim pot that from what I understand should only work if you solder together the "adj" connection, but if I turn it it makes the output go from 4.12V to 4.33V.

Are there some magic tricks I can do to increase the output voltage or is this one of those lessons you learn from buying 0.60€ buck converters?

>>1397449
i tried redoing the soldering work on the 5V connector and now I get 4.98V, so I guess we can confirm once again I fucking suck at soldering. However, after placing a 2 ohm resistor between Vout and ground pins, the output voltage drops to 1.3V. Seeing how this buck converter is rated for 3A, and given that the RPi might actually draw all of those amps depending on its peripherals and computational load, I guess I'm still not too happy.
Should I just spend more money and buy a decent converter?

>>1397518
What's the battery terminal voltage with the 2ohm resistor in place?

>>1397531
I forgot to write it down, but it was a bit weird, I admit I didn't understand exactly what was happening. I used a lab power supply (30V, 5A) as the input to the buck converter. At no load I set the supply voltage to 11.6V or something. When I put the resistor and turned on the supply, its own meters were displaying 4.9V and 0.33A IIRC.
So yeah, I guess now that you mention it it's understandable that the converter's output was lower (it always loses about 0.5V), but then again it still shouldn't go as low as 1.3V...

>>1397170
my turn signals don't even turn themselves off, i'd settle for that before some ultrasonic meme.

>>1397143
yeah i'm retarded. i don't make an effort to be correct though since 5 other people will see to that for me.

>>1397219
that'll probably work

>>1397449
the lesson is to always derate chinkshit by 20-50% depending on item
>https://datasheet.octopart.com/MP2315GJ-Z-Monolithic-Power-Systems-datasheet-26521328.pdf
and also look up the original datasheet, as the other component values may vary by output voltage (especially L1 > 4.7uH for 5V@3A)
>Output Current:3A (max),actual testing input 12V output 1.5A
lel, and also read the item offering carefully
t.got bitten by just that just today, will now be using wire wrap to connect to 2.0mm connectors on nRF51822 "core board"

Which 898D clone hot air rework station should I buy?

>>1396689
I bought a couple of USB diffusers from Aliexpress. They go in a cup of water and make mist by using ultrasonic frequency. One of them quit working completely so I opened it up. Fucking thing was not waterproof at all. Is this shit fucked or is there a chance I could repair it? Pretty irritated about it because I was using it in a sealed 10 gallon aquarium to generate mist to grow plants with.

>>1398084
Looks pretty fucked to me.

>>1398085
>>1398084

Did you submerge the whole thing? Some of them have a bottle or thing that you fill with the liquid to be diffused.

>>1398084
fucked
so now you can buy two new ones and go about waterproofing them before use

>>1398082
the one priced at the 25th percentile. iow, skip the one or two cheapest and anything from a shady looking store, and buy the cheapest one left

>>1397449
looking at the helpful photos provided by ModuleFans, it looks like there's a trace shorting the adjustable-option pins together. do you have to cut a trace?

>>1398093
No, it is supposed to float in a cup of water.

>>1398094
I'm going to go with something else this time. They have some heavier-duty looking ones for a few bucks more. I need it to be reliable so...

>>1398107
>I'm going to go with something else this time. They have some heavier-duty looking ones for a few bucks more. I need it to be reliable so...

go to goodwill and look for an ultrasonic humidifier

>>1398113
That is basically what I've got here. The white disk on the left is the same part in a Vick's ultrasonic humidifier. Vicks obviously is a better brand than IVYSHION. I'm going to go with something like this. I need it to be able to fit in a 10 gallon aquarium.

>>1397804
>actual testing input 12V output 1.5A
Fuck i missed this. Thanks for the pointers.

>>1398096
yeah i did that. Maybe not the best quality job but it's ok.

Well I did it lads, my first fully completed project.
It looks like shit and sounds like ass, but I built it and this one is mine.

I had put an RF and Microwave Engineering class on my resume, because I thought I was going to learn some cool stuff, and it seemed worth mentioning because I haven't met a lot of people who have learned about that stuff. The professor hyped up the class, talking about all the labs we were going to do, project based class, and build impedance matching circuits, and how much we were going to use AWR software and the network analyzer.

Well, none of that ended up happening. I haven't learned shit in that class, except a lot of math. We only got to use a network analyzer once (I go to a small school, so it was the professors own VNA he brought in), and I don't really have any physical/practical sense of the subject other than just solving equations.

Now a company is flying me out to interview with them (had a phone interview with them almost two months ago), and they sent me a schedule and it includes a component where they'll ask me about microwave/rf stuff and use equipment.
Should I tell them I'm not that good with that stuff and send them my updated resume?

>>1398328
What is that shit? Looks solid

>>1398382
Engineering is math. It sounds like you were expecting technician training. Just read up and maybe fiddle with hardware if you have time.

what makes RF Engineering so much more difficult than the other subfields of Electrical Engineering?

>>1398388
Just a simple fuzz pedal for guitar.

>>1398382
You're overthinking it friend. It's not like you wrote you have 20 years of experience in the field. Worst that can happen is you walk away with some interview experience.

>>1398382
Just take I until you.make it, don't worry about it

>>1398390
Yes, but the itinerary explicitly states rf measurement and troubleshooting, which I know nothing about. I just fucked with reflection coefficients and scattering matrices on paper. And I'd love to mess with this stuff on my own time, but come on, rf equipment is $$$$.
>>1398394
>>1398395
Maybe, I'd just feel like a retard if they asked me to do something with a VNA and I just went
>"ummmm"
>*turns knob*
>"uhhhhh"
>*stabs a couple of buttons*
>"........."

>>1396689
Wow, why havnt i found this sub before? You guys are fucking geniuses

what are the job prospects for a BSEE in this day and age? Job security?
I love EE/Computer Engineering and certainly like it a lot more than CS/SE but I'm worried about long term stability in the job market

>>1398411
Look for jobs in your area/country and see for yourself.

I would like to know how to convert 240v ac to dc then some how get that dc in a battery bank then get that dc power from the batteries in an 48v dc input inverter(s)? All without losing as much power as possible.

Would anyone also know how to use power from a generator without putting drag on the generator, what is happening to generator when you power a load off it anyway?

>>1398418
realistically you need to do this in 3 or 4 commercially available modules. you can use any generic offline power supply that puts out slightly above the voltage of your battery bank, then a battery charging board, and then an inverter and maybe a low voltage cutoff circuit. you can just go on ebay or aliexpress, put together a shopping list of those modules, and then ask around to have your mistakes pointed out.

>what is happening to generator when you power a load off it anyway?
energy gets converted from rotational kinetic to magnetic to electric. when a windmill generates, it's literally pulling the kinetic energy out of the air and turning it into current and some heat (inefficiency). there's no way to pull power from a generator without putting some "drag" on it. as a demo, you can short its leads which wastes all its power as heat and makes it act like a brake. what you're probably really looking for is how to extract power from the generator optimally. i don't know what's commercially available but for solar that's called "MPPT" (google).

Howdy br/ohm/s. I got a pickle here that needs uh.. solving. whatever.
I've got a voltage controlled filter notch/bandpass circuit here based on OTAs and I want to take the input current (i 0 in pic) and duplicate it with a current mirror so I can make several notches and bands that all move together by using different capacitor values. But the usual "current mirror" circuit only sinks or sources in one direction; the output of the chip is a current, but it's bidirectional. The output normally goes to a capacitor and the high impedance input of the chip, so the DC bias is all fucked or something. Just inserting a pull-up resistor kinda works; the notch and bandpass curves show up but the level is significantly reduced. I'm assuming this is just because the output is operating too close to one of the power rails, and transient simulation confirms this. What's the proper way to bias something like this? Or is there a way to make a bi-directional current mirror?

>>1398478
I think you can make a bidirectional current-mirror with NPNs and PNPs, though I'm not sure how you'd arrange it in this case. If it were just pulling a current through a resistive load it wouldn't be a problem, but trying to pull a current out/into an IC makes things more difficult. Mess about in spice and see what you can fitter out is my best advice.

>>1398478
>proper way to bias an OTA
all your OTAs are sourcing current. the only dc current sink is the collector of Q2. you might add a pull-down resistor of tens of kilohms, or just a fixed current source, to see if you can get a higher output level from it
also, observe that you're sinking as much current from the feedback node as is sourced from the input OTA, then feeding that ***reflected*** current into the - input of that OTA, which to me sounds like positive feedback. are you sure this is what you want?

>>1398393
Easy circuit models like small signal and low freq models stop applying as parasitic effects easily become overwhelming, for example building a decent VHF amp is bitching because the wire capacitance acts as negative feedback reducing gain and things like P-N junction variable capacitance start mattering

Any suggestions on videos for how to solder, or /csg/ SMD soldering practice kits? And what do you guys use as far as flux/brass wool type cleaner/soldering paste/that sort of thing? I partially blame me, partially my equipment for me being shit, and I want to completely take the equipment out of the equation, both because I'm a gear whore, and because I don't want to ever be able to say the reason I'm bad is anything but lack of practice. My goal is just to get to where I can reliably solder 0805 sized stuff, which is about as small as my projects go currently.

>>1398573
>videos
anything by John Gammell
>SMD soldering practice kits
there are a few. some have rotating lights, some are just boring chains of components. search alibay for "smd practice"
>flux
personally I just use sra-solder's liquid flux pens. tacky paste flux is convenient as it helps keep components in place during work
>brass wool type cleaner
mine came with my iron. as long as it's brass turnings you'll be fine
>soldering paste
you might not need this unless you're reflowing your boards, which is a whole different ball of paste
>etc
get a broad assortment of tips, which should include at least small and medium screwdriver, and a full range of bevel tips. John Gammell recommends a 3mm face-tinned bevel tip as a "real performer". that sounds unwieldy, but it's awesome for doing 2-3 pins of an SOIC at once, or for drag soldering fine pitch QFPs
also get some thin wire solder. 0.5mm is the thickest you should even consider for SMD work. Multicore is the gold standard. leaded solder is easier if you don't care about doing this stuff professionally
and the soldering station? just make sure it doesn't suck. /ohm/ loves Hakko but Aoyue works fine for me

When I'm soldering, is it normal for surrounding components to get kinda hot? I mean I can still hold them but they are definitely warm.
I tend to keep the iron tip (with some tin on it) in contact with the pin and hole for a few seconds to make sure that the solder sticks well to both and goes all the way around the pin.
I'm just a bit worried that I might accidentally fry some sensitive components, I don't really know how much heat they can stand.

>>1398614
>, I don't really know how much heat they can stand.

Devices that are intended to be soldered can handle a lot of heat for a brief moment, as the spec sheets usually say. Pic related is for 7800 series voltage regulators, and that "lead temperature" is for soldering purposes.

>>1398618
ah thanks, I didn't know it was there.
>260C
I guess if I can still touch the components/board without it hurting it means I'm nowhere near dangerous levels of heat then

>>1398624
I've burnt components to death before, but that's in my early days before I discovered the age-old method of "shit ton of flux" and kept trying to get a good joint without letting the part cool down.
But sometimes it's necessary to heat up the board you're working on, usually if it's got a lot of heat-sinking nearby like an aluminium-backed COB. An upside-down clothes iron does the trick, or so I'm told.

>>1398418
Look up the specs of the generator. When you increase the load the fuel consumption of the generator increases as well. This is where you get the power from. Ideally you'll want to run the generator at maximum efficiency, which will be in the upper load range, or maximum load. Otherwise you're just wasting fuel. Whatever converters and what not you're running after the generator should have a peak efficiency at this load.

>>1398633
I am using lead-less solder with flux core, and my iron is set to about 300C. I think the joints are coming out ok considering also I'm still quite inexperienced. Sometimes I still make some minarets or unspeakable atrocities, but I think I'm improving.

There's one thing though that's bothering me and I'm not sure how to fix it or if it's even important, but quite often i get these brown stains around or on the joint. I think it's the flux burning but I don't know if I have to turn up the heat or lower it...

>>1398651
use 99% isopropyl alcohol and wipe it off. flux isnt needed after its cooled down. use microfiber cloth also.

>>1398658
I use an old toothbrush and isopropyl, rags/wipes etc tend to catch on the pins.

>>1398614
You may not be using enough heat. Seems counterintuitive, I know, but you should be able to get in, make the joint, and get out in 3 sec or less. By making your joint that quickly, heat doesn't have much of an opportunity to transfer to parts of the board/other components that you're not trying to stick solder to.

>>1398651
300 might be a bit low, I run 350C with a 30W iron, 300 gives me cold joints. 400 leaves me with great joints, but the flux smokes off too quickly, to say nothing of heating up the PCB too quickly.

>>1398658
>>1398676
I use a toothbrush and IPA, but I find the toothbrush doesn't get rid of all the flux so use a tissue to clean the rest. The tissue usually catches on the pins and leaves tissue fragments. Is a microfibre cloth easily cleanable of flux or not prone to catching?

>>1398658
>>1398676
>>1398687
>>1398701
thanks for the tips. Isopropyl alcohol was already on the shopping list because I'm actually having the same problem on my iron tip. After some usage it becomes darker and the tin doesn't stick to it as much (in the worst cases it doesn't even melt, but the heat travels up the wire to my fingers), and I read wiping it with alcohol can make it as good as new.

>>1398711
Sounds like a problem fixed with flux and copper wool, not IPA. A replacement tip may be in order if it's pitted. Just tried spraying IPA on my tip, it cleaned off some residual flux but it isn't that much shinier.

If you have trouble getting the solder wire to melt but it's heating up the solder wire you're definitely at too low a temperature, or your iron has a very low heat output.

>>1398711

very brief video of it working:

https://www.youtube.com/watch?v=WFaOPb017FM

>>1398715
>>1398716
>temperature
I've set it to 400 now, I guess we'll see next time I solder something. I'll also try 350 and see how it compares. For the rest I just need to visit my local hardware store.
Thanks again.

>>1398716
That shit's like $15

>>1398729
>That shit's like $15

Some people actually have money.

And even if you don't, raise a loan somehow, and buy it. Even if you need to use it every year or so, it will last a lifetime, and you can bequeath it to your eldest male child.

How do I figure out the minimum voltage and current of my circuit, and the maximum that it can handle?

>>1398757
the obvious answer is that your circuit's voltage range is the maximum value of all your parts' minimum voltages and the minimum value of all their maximum voltages. i don't think you'd be asking if it was so simple, so what are you making or abusing?

>>1398758
I'm circuit bending shit and don't wanna blow shit up if I can help it

>>1398759
If you're circuit bending you better leave that fear at the door, son.

>>1397055
No if you add some resistance to add to 4 ohm.

Are ther any schmidt trigger logic ICs out there that can take 12volts while the IC supply is only 5V?

wireless relay board

Has anyone else here ever found themselves in a situation where you need literally just one single part and there's no hardware store near you that has it, so you'd have to order it online but you don't really want to because it would be such a waste to make the postman come all the way to your house with a single piece of electronics?
What's the solution I'm not seeing?

>>1398907
>wireless relay board

looks positively ghey.

>>1398914
>What's the solution I'm not seeing?

more cabinets

>>1398914
yes but I work at an engineering firm that will sell me parts from their stock.

>>1398914
order 10 of it on ebay for $1. eventually it'll stop happening.

>>1398711
A wet sponge/rag will usually solve this problem.

>>1398701
you need a lint-free tissue, such as Kimwipes

>>1398722
375° is about optimal for everyday soldering such as pin headers, balancing ease of use vs. tip life. turn down 20° or so for smaller welds, up 20° or so for larger welds

>>1398729
it's $15 well spent (I only paid $8 on fleabay for mine). seriously, that little tub will last approximately years

>>1398864
probably, but I don't know of any. use a resistive divider

>>1398907
>low voltage signals prancing right between line and neutral with the bare minimum of 1.5mm between mains and low voltage
>skinny-ass traces for 30A
fire hazard/10, you're gonna need 20mm wide traces at 2oz copper for 30A, or some channels in the mask to lump solder onto, or your artwork on both sides of 2oz board. also double your clearance between mains leads and between mains and low voltage
may I suggest you try again with the ac-dc converter at the bottom and maybe a 12F series PIC

>>1398914
moar stock

>>1398942
it'll cause other problems, like faster tip wear. brass/copper wool isn't just a meme

>>1398969
the board is two identical sides with mask gaps on the bottom. clearance is a function of space constraints to an extent but i'm happy meeting the iec clearance requirement given the environment and that the relay coil traces can function with some leakage and up to one bridge.

>>1398916
What do those stacks of cabinets cost? Been meaning to get some but I'm not sure if I can justify an individual drawer for each resistor value. I'm thinking a bunch of (angled) pieces of plastic/aluminium pipe glued together on a slab of MDF or plywood instead for those small parts.

I'm using only 12 volts but the PSU can pump out 1200 W. As the wire length is only 1-2 meters, should I be worrying 2 mm thick copper wire causing fires?

How can I make this circuit medical grade safe? I made this circuit based around the INA128 instrumentation amp, and it seems to work pretty well when tested with my oscilloscope. However, I would like to make a full ECG with a VGA monitor as the output display; doing this with the confidence that it won't kill me if something fails would be desirable. Does anybody have any resources that could help me?

Additionally, I have omitted some information; I'm mainly posting the circuit just for the background.

>>1399075
one decade per container is pretty manageable for E-24 and smaller series, depending on quantity

>>1399100
isolating your sensors galvanically from the rest of the device is a good start. that implies an isolated, clean dc-dc converter, and also weakly implies putting the ADC on the sensor side of the isolator, with all your video crap on the line side. the VGA monitor is probably the weakest link
also consider a brick type power supply, which can be procured in many levels of isolation capability as suits your comfort rating
>medical grade
between safe and "medical grade" there is mostly a shit-ton of paperwork. I suggest you put a few kilohms on the inputs and call it a day

>>1398992
>meeting the iec clearance requirement
But not creepage requirement. Or at least it doesn't look like it.

>>1399100
>Does anybody have any resources that could help me?
IEC 60601-1 for Medical Electrical Equipment Safety
https://industries.ul.com/healthcare/medical-devices/ul-60601
Electromagnetic Compatibility Engineering (book) has sections dealing with noise.
Use a medical grade power supply with your design.
You should optically isolate the VGA signal. Might even be able to get something like that off the shelf, you can for HDMI/DVI.
You can protect yourself further by not using wall power at all, and operating only off batteries so that there is no danger of zapping yourself.

>>1399100
Like the other anon said, galvanic isolation would be a good starting point. The isolation from the inputs to shit like your computer is required to be worth of reinforced (double) insulation.
Medical shit is also required to remain safe in case of a single fault, which would require (at least) input current limiting resistors.
While these aren't related to safety, your input stage has too much gain to be reliable, you should add good overvoltage protection and your amplifier inputs need to be biased one way or other.

>I have omitted some information;
This rarely leads to anything good.

Why are people in the eevblog video comments so insufferable?

Okay so this will be a really retarded question but i really don't understand electronic well at all.

So, I have raspberry PI and want to connect some things to its GPIO pins. Basically a pin is just a metal stick with 5V in it.

So, let's say I have a LED light with two legs that needs 3V to function.

So if i do

5V pin ----- 3V LED ----- GROUND pin

The LED will burn

So i need to do

5V pin ---- 3V LED ---- 2V resistor --- GROUND PIN

Simple enough... but here is what really confuses me.. it's not just volts, it's also amps, so the pin supplies also amps along with volts and the LED also needs certain amps and not just vols... but what are the amps then? if the LED says it needs 3V then why do i also have to care about amps?

When you done laughing at me please explain it to me, since i don't want to destroy my Raspberry. For some reason my brain seriously struggles with this.

>>1399123
>utoob comments are retarded
Stop the presses, all of them.

>>1399125
You need to care about the LED's current rating, because the current through the LED increases rapidly (to destructive level) after you exceed its normal operating voltage. It isn't entirely impossible to run LED from a constant voltage source, but the manufacturing tolerances, temperature variation etc. make it usually impractical.

Your resistor isn't a "2 volt" resistor, it is a resistor which drops the extra 2 volts when some suitable current goes through it. You can choose whatever current, as long as it is less than the LED's maximum rating; this controls the brightness.
Another way to see it would be that your LED is a current-driven component and that your resistor converts the 2V voltage difference to drive current.

>>1399125
5V - (3V LED) = 2V = (LED current)*(resistor ohms)

Find the LED current from the datasheet for the part. It can be hard to find exact resistors by the math so as long as the current is less than the max on the datasheet then that resistor will work fine.

>>1399125
the pin supplies volts.
The led draws amps.
Resistor is measured in ohms.

Ohms = volts / amps
so pin is 5v, led is 2v, 3v left over.
Led needs 20mA, or 0.02amps, check the datasheet it will tell you current draw.
Resistance = voltage / current
R = 3/0.02
Resistance = 150ohms.
That's what resistor you need to burn the extra voltage at the required current.

Also, too much current will kill the led AND the pi so use a resistor is very important.

The formula used today is called 'ohms law's have a google.

I just realised i got the voltage wrong but I'm not going to change it, do the calculation yourself and post results like homework

>>1399131
Yeah but everyone in his comments are super self proclaimed experts with unlimited knowledge.

>>1399137
>led is 2v, 3v left over.
you scrambled the values mate

>>1399142
Read the whole thing, I was having a dump and posting from a tablet, i don't hate myself enough to go back and change it and my legs were getting numb i didn't have time to redo it

>>1399146
I posted before I finished reading your whole post but I'm not going to change it, do the calculation yourself and post results like homework

I was having a dump and posting from a tablet, i don't hate myself enough to go back and change it and my legs were getting numb i didn't have time to redo it

My friends and I finished a sumo robot.

It's pretty shit but it was fun and taught us a lot.

>>1399192
Do i smell a chink 3d printer?

>>1399194
We got the chassis printed at our schools printer, which is free for students.

>>1399103
as i understand it the creepage requirement is superseded by the clearance requirement for pollution level 1. this goes in an ip65 enclosure so i'm not concerned about condensation or much dust. i have considered the thermal implications of that, but only with rough math.

Got a lecture on electromagnetic interference the other week. Kinda scary to be honest, the guy told us about a person who was hospitalized and a relative made a phone call while close to the medical equipment. The EM waves from the phone interfered with the controller of the morphine(?) pump and the patient overdosed and died.
Another example he gave was some dude walking under a suspended load and some EMI made the crane drop the load. rip.
With more and more electrical devices being installed and with the fact that these devices are using lower and lower signal voltages, it was clear why it's so important to shield your stuff.

>>1399240
I'm guessing that the cell phone/ infusion pump example was from the 1980s or something, that seems highly unlikely now as you mentioned signals frpm all of our mobile devices are much lower power. And at least there are standards for medical devices for EMI- resistance.

>>1399245
Yeah probably. Maybe not all the way back to the 80s but definitely before they started implementing standards and rules. There are limits to both EMI protection as well as emission

I'm confused as to how capacitors work

Do they only charge when the circuit is open?

When the circuit is closed, and the capacitor discharges, does it just continue to let current flow through it as if it weren't there?

>>1399288
Capacitor charges until the voltage across it is the same as the voltage source being connected to it.

>>1399107
>>1399106
>>1399101
Thanks, I will definitely look into your suggestions.

I thought car relays were intended for staying energized for long periods of time.

I'm unsure what the chink mean with "Max operate voltage". Is it just Chinglish and did he actually mean the voltage required to energize to coil and switch the contacs?

Or should I step down the voltage if I want to leave the coil energized.

https://www.aliexpress.com/item/Car-Truck-Motor-Automotive-high-current-relay-12V-200A-2-4W-Continuous-type-Automotive-relay-car/32534150838.html

>>1399448
/o/ here (yes, the whole board)
My guess is that "max operate voltage" is the limit before the contactor welds itself open
Car batteries have a shitload of amps for starting

Say I need to provide +5, -5 supply for my instrumentation amplifier

Do I use linear regulators or voltage references, or both?

I'd need two different regulators, right? Can't do it with two +5 regulators?

Noob here
How do i know how much volt an opamp had to correct with?
+ pin: 5v
- pin: 4.4v
Out: ??

>>1399297
What about ac?

>>1399520
uhhh what?

>>1399448
"nominal voltage" is what you should drive it with. but relay coils can actually actuate the contact at a lower coil voltage than the nominal one. given the context i'm 90% sure the "max operate voltage" is really the *minimum* coil voltage at which the relay will operate.

note that the next column says "mix release voltage" so the translator was clearly out of his mind on opium.

>>1399519
>Do I use linear regulators or voltage references, or both?
you can use linear regulators. the dc component of the supply voltage isn't critical. any ac on the rails will be attenuated per the PSRR datasheet spec. for example the INA126 has a ~95db PSRR so a 100mV 120Hz ripple on its supply would be reduced to 1.8uV on the outputs.

Note: Sorry for incorrect grammar.

Are linear components, the ones that require heatsink any good or should i avoid? everyone says use switch mode types but if i got thermal paste and correct heatsink why not use linear? yes i know its not efficient but anything else? same as resistor based led drivers, besides power efficiency and if i mitigate heat, why not?

another thing... when you short a capacitor to discharge it without resistor etc is this damaging the cap?

thanks /diy/ learnt so much from here.

>>1399520
read the datasheet
normally, the output of an opamp is the voltage difference between its inputs, multiplied by the first parameter of interest, open-loop gain, Av (usually in the several tens of thousands), limited by the second parameter of interest, its output voltage range (for common opamps, very close to the - rail on the one end, ~1.5V short of the + rail on the other)

>>1399644
1. it sounds like you know the trade-offs well enough
2. the failure mode of concern is I^2*R (heat) damage. generally speaking, the larger the capacitor and the more complete its charge, the more likely the crowbar discharge method will degrade or destroy the foils etc. inside the capacitor

>>1399644
In addition to what the other anon said: if you have a polarized capacitor and the capacitor and the discharge wire are inductive enough, the resulting reverse voltage might kill the capacitor.

>>1399644
>Are linear components, the ones that require heatsink any good or should i avoid?

it depends on the current being drawn and how much you need to lower the voltage. say you're driving some op-amps which draw 30mA, and you're dropping 3V (from 8V to 5V). so the power used by the regulator is 3*.03 = .09W or 90 milliwatts. this is nothing, so you can use a tiny plastic package without any heat-sinking at all.

also, linear regulators provide very clean power, unlike switchers, which is important when dealing with tiny signals.

Are the cheap (~20c ea) SPDT panel-mount switches on ali decent quality? Been buying all my switches when I needed them locally, but I think it's time to get a little stockpile since I use them for damn-near everything.

A source of small, sturdy (~1mm wall thickness) metal tins to mount them in would be pretty nice also but I'll not get greedy.

Does /ohm/ have a recommended shop for electronic components? Something with a diverse stock, good service/quality, etc.

>>1399828
Are you retarded?

>>1396689
>Components/equipment sources?
Mouser, Digi-Key, Arrow, Newark are global full-line distributors with small/no minimum order.
eBay/AliExpress sellers, especially good for component assortments/sample kits (caveat emptor)
Independent retail electronics distributors
ladyada.net/library/procure/hobbyist.html

>>1399839
https://www.elfadistrelec.no/

They sell to a lot of different European countries.

Pricey, but they do carry a lot of industrial and you can buy it as a consumer.

>>1399839
>Are you retarded?
Guess so, I forgot to specify I meant aliexpress shop

>>1399846
no alishop carries everything you might possibly need, of course. I usually build an order from a handful of shops of at least $10-20 each when I go component shopping
shops I have dealt with happily and more than once who ship quickly enough and don't fuck up: Parity supermarket, Chinese Super Electronic Market, Survy2014 store, Striveday, UltimateElectronics
highest honors for ElectronicFans which has their shit together and ships quickly. I regualrly receive shipments from them inside of 2 weeks in the USA via ePacket
things I look for include: photos match product, cart automatically combines shipping for discount, seller doesn't mis-describe product. don't shop too too hard on price. you do get what you pay for, even from china

>>1399895
>even from china
probably especially from china

How do you attach your eletro stuff to things?
Like i have arduio with som relays and other shit and i can just drag it around by the wires.. how do you solve storage issue?

>>1399957

everyone now owns a $5000 3D printer for making our own cases. those who have a job also have $10000 laser cutters so we can work with metal and wood as well.

>>1399957

everyone now owns a $5000 3D printer for making our own cases. those who have a job also have $10000 laser cutters so we can work with metal and wood as well.

you should see our collection of dragon dildos.

>>1399957
arduino and wires is for prototypes or projects you don't care too much about. Once you've done your little breadboard experiment and everything works you design your pcb and send the schematic to your favorite printing company. Get your SMD parts delivered and start soldering.
This guy shows you the complete process in three videos, just skip ahead if you're getting bored, it's not necessary to watch it all. https://www.youtube.com/watch?v=0D5lcNIEa24

>>1399957
You can also get an enclosure and some terminal blocks.

Has anyone here ever used the l298 motor driver board from seeedstudio? I have a hand-me-downs from my late uncle and the datasheet for it doesn't say anything about what the jumpers do or how to use the current sensing pins.

Does antbody know if there is some kind of coating on dip package pins, or is that stuff you see just oxedasion. I just got an ic pretty deep into my foot and I am wondering if something dangerous might have just entered my blood stream.

Say I have a differential signal and I need to scale it down a bit, how would I do that?

I cant use a simple resistor voltage divider since I can't touch the ground

>>1400099
no coatings, bacteria are the only thing to worry about

Hey guys, absolute newcomer here, never touched anything outside my study lab. Can you guys help me out on how to find some kind of product keys for specific parts? Mainly the grey capacitor and the brown resistor in pic related. Like just googling C/M101 brings nothing on my side.

>>1400164
Second pic.

>>1400165

BTH330 appears to be a PTC thermistor of 33 homos, while the brown thing is ambiguous: it looks like a .001uF cap at 100V, but the marking says F4 which is used for fuses. measure resistance across it. if it's zero, its a cross-dressing fuse.

>>1400165

BTH330 appears to be a PTC thermistor of 33 homos, while the brown thing is ambiguous: it looks like a 100picoF cap at 100V, but the marking says F4 which is used for fuses. measure resistance across it. if it's zero, its a cross-dressing fuse.

>>1400189
I'll do that, thanks!

>>1400130
assuming the common-mode voltage is okay and you just need to reduce the differential voltage, use two simple resistor voltage dividers and disconnect ground

Assuming I've potted the connection properly, does this type of wifi antenna need weatherproofing to be put outdoors?

>>1400256

water can obviously leak inside which will rust the wires. given that it's most likely copper, it may last a few years as long as the antenna is pointed up so water will not get inside too easily.

>>1400256
Making a cantenna/wok-fi?

>>1400164
Those brown things are Murata's (CM) inductor-capacitor(-varistor) combinations. Probably 100pF (101), 100V. Dunno what type exactly, but you can try your luck with https://www.murata.com/en-us/products/emc/emifil/lead

>>1400269
not down?

>>1400256
the (RP-)SMA connector is the weak link here, I think. but I would probably just get rod/tube and build an antenna that stands up to weather and provides the desired radiation pattern

>>1400327
>not down?

how is moisture gonna run off if it's pointing down? but if you dont need the angling mechanism to get a better signal, i suppose you could just caulk the opening at the bottom and make it stay at a fixed angle. and thus waterproof.

another concern is the plastic shell tho. it's likely they used some cheap plastic that'll become brittle in the sun after a few months.

>wire calculator photovoltaic cells in parallel
>low voltage but amperage is boosted
>wire output to joule thief
>now have boosted voltage and amperage
>utility has now increased many-fold
am I thinking this through correctly

>>1400381
>amperage
every time

>>1400385
okay, current

I have a 9.6V 1Ah NiMH battery for a cordless power tool but without a charger. To my knowledge the charger ran at 12V 0.5A output. To charge the battery do I have to purchase a charger with identical output specs? Or is there a bit of leeway? Will it charge at 18V 0.5A output? Or 12V 3A output, for instance?

>>1400396
I think if you tried to charge a power tool battery off of .5 amps you'd be charging it for years

>>1400400
For a cordless rotary tool.

>>1400381
No that's a fucking perpetual motion machine.

>>1400400
It's not that bad, it should be done in a day or two.

>>1400396
The charger would have been limiting the voltage into the cell, just feeding it a 12V-max DC would overcharge it and probably leave it right-buggered. I'd google charging profiles for NiMH cells and see if you can match one with simple circuitry. But trickle-charging is usually always an option.

>for years
>in a day or two

0.5 A * 3 h = 1.5 Ah = more than enough for a 1 Ah battery

what's going on here?

>try and de-solder components from old motherboards, various old PCBs
>solder won't even melt, even when I turn my iron all the way up
>when it does melt, the solder wick doesn't absorb any solder
>even if I flux everything to hell and back
what gives
does solder just get harder and harder over time?

>>1396689
i went on a road trip and i still have to go back and wanted to upgrade my car audio system. i checked the local craigslists and found some subwoofer+amp combos for around 200$. can I hook this up to my stock reciever and speakers? is there anything I should look out for or keep in mind? thanks

>>1400411
1Ah? I was assuming something closer to 15-20, though now that I think about it that only makes sense if they're all in parallel.

I want to wire up this 120VAC motor that I do not have the wiring diagram for.

I've seen other motors where the brown wires are connected to the start capacitor, but one of my brown wires has a connection piece that doesn't fit onto the capacitor's pin, only another wire.

Any help getting this motor back to working?

>>1400416
in automated processing they do use higher-melt solder sometimes, but the effect shouldn't be that extreme, too small a tip? bad contact angle? not heating the solder through the wick? need a more active flux?
>not using hot air for component salvage
son, I...

Is there such a thing as a dummy battery?

I want to do image related, but don't wanna solder directly to the terminals.

So if i have say a light bulb that needs 10V and i connect it to a 100V power source, then i add resistor to eat the remaining 90V, what happens if i want to connect another light bulb that needs 20V to shine at full power?
Do i just add resistor that eats 70V and then they will both shine at full brightness without the 10V bulb burning out?

>>1400500
This sounds like a school physics problem. Why don't you check google?

>>1400515
I am just beginning to learn this stuff so it confuses me.
For example if i have a bulb that needs 20ma going through it to operate, then why does it also list voltage? say 3v/20ma I can adjust any voltage with resistors to produce 20ma, so why is it important to list the voltage?

>>1400498
yes, there is, but y tho when you can make one out of a couple of button head screws and a length of wooden dowel

>>1400520
the relationship between voltage and current through the lamp (or other load) may not be linear. you need to deduct its voltage rating in order to set the current limiting resistor correctly (see Kirchhoff's voltage law)

>>1400523
So if i have 10v power source, some resistors and stuff, then by the kirch law the node before the light bulb to be say 5V, since voltage drops going thru stuff in the circuit, then that means that to calculate resistor value for a 20ma bulb, i now have to use 5V inside the ohms law instead of the original 10V from the power source, do i understand it correctly?

Anyone know which NTC 10k curves the are based on?

Read a little bit about NTC 10k sensors, and the temperature to resistance scale vary a little bit between them.

https://www.veris.com/docs/support/faq/RTD-Thermistor_Z202030-0N.pdf

>>1400396

chargers are made to closely match the battery. specifically, the transformers are sized so they limit the current to some safe value. if you just hook up any old power supply, you run the risk of overheating the batteries. if you have instruments, then you can use power resistors or light bulbs in series with various transformers until you get a safe current of around 0.15A.

>>1400416
>does solder just get harder and harder over time?

not really, but adding new solder helps. if dealing with motherboards, you usually have a ground plane that sucks up all the heat. sometimes i use 2 irons at the same time to overcome that.

>>1400498
>Is there such a thing as a dummy battery?

there are adapters to convert AAA to AA. or, for the negative side, you can use an alligator clip, and for the positive, a dowel cut to size with a washer screwed-on will work.

just accidentally shorted the 2 deflection coils on my crt oscilloscope
for reference one runs at 50hz-ish and the other at 15hz-ish
smells like something burnt so im assuming its done for? cant find anything on google

>>1400667

find schematic and test/replace the transistors that drive the coils. if it's a chip, it's gonna be harder to find and to replace.

>>1400269
Thanks dude

>>1400296
Yagi actually. There's a printable one on Thingiverse you just pop onto the antenna after placing copper rods in it. Initial tests are going far beyond my expectations. Never heard of Wok-Fi though, I may test that out for longer-range applications (Although I'll prob just skip that and go straight to lorawan, as I'm only sending a few bytes at a time via mqtt)

>>1400327
Ive got shittons of these little antennas, if I didnt I would do just that

>>1400327
Yeah itd prob be prudent for me to just cover the whole bottom half with liquid electrical tape. Should I also put grease on the parts where the connections are to repel water?

>>1400449
Measure the resistance between a bunch of the wires until you figure out which pair of coils make the start-winding and which pair make the run-winding. There are only six wires instead of the maximum of 8 for a 2-phase multi voltage, so I'm guessing that instead of running two different winding configurations for higher/lower voltage for the start-winding you just use a different capacitor.

Not sure how you'd determine which way around the windings need to be wired, trial and error is my best bet.

>>1400500
I'd throw them in parallel with a resistor for each bulb, unless they happen to require the same current. If you're not given information about the current the bulbs draw / the resistance of the bulbs then just assume the currents are different and throw them in parallel. But if they do pull the same current then it's no problem to put them in series with a resistor made to drop 70V.

Is there a correct way to strip tiny wires? I've been holding a lighter under them for a couple seconds and pulling the insulation off, I assume this is a retarded way to do it

>>1400774

tiny stripper. nail clipper. teeth.

>>1400774
Thermal stripper.

>>1400774
I find a pen-knife works well.

>>1400532
you understand it perfectly

>>1400602
not sure which curve that table is on but iirc B3850 with rate value @25°C is a common curve for temperature sensing. the only way to know for sure is to calibrate

>>1400683
>grease
I wouldn't, just leave it be air
>liquid electrical tape
dude, you gotta try some of that self-amalgamating tape. it's a life-changer

>>1400809
>you understand it perfectly
I can't tell if you are being sarcastic or not

>>1400809
>self-amalgamating tape

I had no idea this shit existed. Gonna grab a roll in the AM and wrap my shit up before I deploy it tomorrow. This liquid electrical tape seems kind of shitty, you probably just saved my ass. Thanks brej

What do I need to know about charging ni-mh solar light batteries?

>>1400823

plus to plus, minus to minus.

>>1400813
not at all. the resistor of interest itself is in fact the correct V to figure R=V/I in a series circuit

>>1400823
ideally, use a purpose-designed charger IC with automatic charge detection
less ideally, charge at no more than C/10 with a 12-hour charge timer

I hear you can get broken oscilloscopes pretty cheaply (or freely) and then fix them to get a good scope on a tight budget.
How the fuck do I find broken oscilloscopes, and how the fuck do I go about fixing them?

>>1400906
>how the fuck do I go about fixing them?

Quite often you will need a scope that works in order to diagnose a broken scope.

>>1400910
Are there any common places that I can skeeze my way into to use a scope? I suppose I'm young enough to look like I'm supposed to be in the electronics lab at a university.

>>1400910
Maybe a $20 DSO138 would be good enough?

I want to get your expert opinions on something.

So, I have this 3D printer. But, because I am exceedingly poor, I had to buy one without a heated bed. The bed is the thing that moves and the printed object stands on it.

What I would like to do, is to buy a heated bed part from another printer and mount it on, instead of my current shitty one.

But here is the problem. I need to power it and keep it at a constant temperature of 100C.

How hard it would be, to power the heated bed from a regular 230V outlet AND lower or increase the el. current to keep the bed at a constant 100C temperature? It is okay if the temp wobbles around +-10C

>>1401000
What's your budget?
Off the shelf solutions exist (PID temp controller + probe) but they aren't "dirt cheap" which is what you're looking for I bet.
If you do it yourself with components it's pretty cheap, like a few bucks.
google: 555 temperature controller circuit triac
Or generally: "opamp temperature controller circuit"
Basically what you'd be doing is building an analog thermostat with a low-temp setting and high temp setting, thermistor as the sensor.

>>1401000
>>1401010
Actually don't the heated beds come with their own controller?

>>1401000
which printer did you buy? most printers have standardized parts and control boards that are designed with expansion in mind, e.g. extra fans and whatnot. So it could be that although your printer was sold without the heated bed, if you buy one especially made for 3d printers it might be possible to just plug it into a port on your control board, along with its thermistor (also shipped together normally). Then it would just be a matter of enabling it in the software (again standardized and comes with PID temperature control).
Did you try asking in the 3d printing general?

Can someone take a quick peek at this instrumentation amplifier

http://www.ti.com/lit/ds/symlink/ina128.pdf

If I'm using a Vref that ISN'T connected to the ground, do I connect bypass caps at (V+,GND),(V-,GND) or (V+,Vref),(V-,Vref) anyway?

I'm thinking the former but I'm not quite sure anymore?

Page 19 includes layout example

>>1401042
tevo michalangelo

>>1401010
the beds i found cost around $40 i don't want to go over 50

>>1401014
I don't know. They definitely have to have temperature sensor so the printer can regulate it but i imagine the temperature regulation itself is done by the printer which changes the current going into the bed

>>1400396
Would one of these Turnigy Acucell chargers work to charge my rotary tool battery? Just set it to 12V 0.5A and connect to positive and negative? If so, I might get one.

Another idea, is it not possible to open up the battery and charge the individual cells? I just looked and there are eight 1.2V cells. Can I not just put them into a AA charger and package it all together again when they're done?

>>1401063
i bought one of these to charge my 3S LiPo and as far as I remember it could indeed charge pretty much any chemistry, with selectable voltages and user-input currents.

>>1401051
(V+,GND),(V-,GND),(Vref,GND)
data sheet tells you why

God gave me this power

>>1401063
I they're 1.2v ni-mh then I don't see why not..

>>1401272
I'd have just sanded off the solder mask by the broken traces and soldered a small copper wire or two before gluing the PCB back together. What's the board from?

>>1401421
>What's the board from?

kind of obvious it's one of those driver boards that plug into the back of a CRT. this is probably the same dude who blew up his scope. >>1400667 and who doesnt seem to realize that rubber bands break down fast if they're heated often.

>>1401464
Band might be there just to hold it while some glue dries.

Anyone know what this symbol on many electronic devices represents?

>>1401523
https://en.wikipedia.org/wiki/UL_(safety_organization)#Recognized_Component_Mark

>>1401525
thanks. what a weird logo.

If i want to connect a 3v led to 230V outlet and use the correct resistor, will it melt because of all that voltage?

>>1401542
If it's big enough: no (20mA @ 230V = 4.6W, 1mA @ 230V = 0.23W). You need to protect the LED against reverse polarity, though, and at higher currents it usually makes more sense to use a series capacitor to drop the bulk of voltage. If you use a dropping capacitor, you need an additional surge limiting resistor.

>>1401545
>You need to protect the LED against reverse polarity,
Unless you solder it in the other way round this shouldn't be a problem no?

>>1401523
it means the Russians are up in your electronics, my dude
>weird logo
yeah, it kinda has to be to enforce the exclusivity right of trademark

>>1401566
your 230V outlets are dc? cool

>>1401566
Mains voltage is AC. Your LED sees around -300V peaks 50 (or 60) times per second.

Simplest way to protect the LED is to put a diode in reverse parallel with it. You can also use another LED in a similar manner, or you can use a bridge rectifier in front of it, which has the additional benefit of doubling the LED blinking frequency to less annoying 100 or 120Hz.

>>1401573
I don't think that's what he was asking dipshit
people like you are cancer

>>1401573
>your 230V outlets are dc? cool
Not everyone lives in a third world country with AC outlets anon..

>>1401574
>Simplest way to protect the LED is to put a diode in reverse parallel with it

no, that's not the simplest way. it means you're using power on both the + and - cycles, which means your series resistor has to be higher power. putting the diode in series is a lot smarter.

>>1401574
>Simplest way to protect the LED is to put a diode in reverse parallel with it

no, that's not the simplest way. it means you're using power on both the + and - cycles, which means your series resistor is getting twice as hot. putting the diode in series is a lot smarter.

>>1401623

on the reverse cycle where the peak voltage is 300+ volts, how do you know how much is dropped across each of the three elements? the reverse current is tiny, so ignore the resistor. the reverse characteristics for the two diodes determines how the 300+ volts is shared.

tl,dr, how does that protection diode protect the led?

>>1401623
Series diode is only simple if you've got a free ideal diode. With antiparallel you just need whatever diode with blocking voltage >= LED forward diode and current >= diode forward current set by resistor, don't care about leakage, reverse recovery, capacitance.

With a series diode it'll need to be rated for the same current, but now has to be rated for full mains peak voltage, and has to be pretty quick recovery to make sure it's off before the mains waveform hits LED reverse voltage rating, has to be very low leakage so that it's the one dropping most of the voltage, not the LED, and has to be pretty low capacitance too.

Your dropper resistor is going to have to be quite a bit higher power rated than antiparallel with dropper cap since you have to have all the mains voltage across the resistor rather than having most of it just across the cap.

So I need to build a motor speed controller with these specs:

Operating voltage: 36V
Maximum Current Output: 15A (But will rarely need to supply more than around 12A)

Maximum input voltage: 40V
Minimum input voltage: 0V (or whatever, it will never reach below ~30V anyway)

PWM controlled (via 5V PWM signal from a microcontroller)

That's about all the specs that are absolutely necessary. I've been doing some hunting and so far the best guide i've found is this one:
http://www.instructables.com/id/DIY-MOSFET-Motor-Controller/

This guide, however, falls short of my voltage requirements. I'm an electronics engineering student, but unfortunately I don't have any direct understanding of these circuits so i'm struggling here.
How can I change the above guide so it meets my voltage and current requirements? Or if that's too big of an ask, what should I research in order to find out?
My assumption is that finding MOSFETs better suited to the voltage and current ranges I need would be the solution I seek.
How wrong am I?

>>1401628
>the reverse characteristics for the two diodes determines how the 300+ volts is shared.

those characteristics are well known: the LED will break down very early but the diode will not break down at all, so it blocks substantially all the voltage, and it lets only a few uA leak through, thus very ably protecting the LED.

>>1401661
Do you need to reverse the direction? Maybe i'm wrong but I think a simple H-bridge would do.
https://www.pololu.com/file/0J52/vnh2sp30.pdf

>>1401661

is he limiting to 30 v because of the relay coil? If so you can work around that by either not using a relay at all, or driving the coil with some other voltage that you will have to provide.

IRLB3034 might work for you.

>>1401661
Looks like you need a buck-boost converter seperate to your speed controller, or a boost at the very least if you plan on powering it with 30V. What sort of motor is it? For 15A @ 36V it shouldn't be that hard to get yourself some appropriate FETs, octopart is your friend if not digikey or mouser.

When selecting a replacement FET you'll want one with a sufficient maximum voltage and current ratings, lower or equal R_DS(on), lower or equal Q_gate, and a V_gate_threshold significantly below your PWM voltage. Might be missing something here.

>>1401690
Those are some nice specs.

>>1401694
>Those are some nice specs.
Yeah, if by nice you mean bullshit. Your mosfet and its wires remains solid only if you use it in very short pulses.

>>1401700
Well I meant the R_DS more than anything. Not like he's going to whack some sort of liquid nitrogen cooling on his FET to run it at 200A.

>>1401686
Reverse would be nice but not at all necessary.

>>1401690
Your coil suggestion may be on to something.
Like I said, I don't have direct experience with this type of circuit, so could you, or anyone really, explain what that coil actually does here and if its actually required for full operation of the speed controller?

>>1401694
I'm not sure about a buck-boost converter.
I plan to use a 36V SLA battery (which actually outputs between 38 and 34V depending on charge) to power a 250W motor.
At these specs, the 36V 15A output will be overkill for the 250W motor since it won't be able to sustain that much power for long periods of time. That, however, isn't a concern. I will code in power restraints on the micro controller.

I don't really understand your recommendations for selecting a replacement FET, but thank you anyway :)

>>Those are some nice specs.
So would replacing the original guide's MOSFETs with IRLB3034's and replacing the relay coil with a 36V+ one work? Or perhaps coild I remove the coil altogether?

>>1401700
>Yeah, if by nice you mean bullshit. Your mosfet and its wires remains solid only if you use it in very short pulses.
This guy seems to have other ideas. So what would you recommend?

>>1401704
Coil is only needed to be able to reverse the motor. If you don't need to then just a single FET and one diode would be fine. Maybe a resistor or even a transistor on the gate, for reasons.

Will you have a current shunt on the line in to monitor current and a divider to measure the power used by the motor? The motor's resistance will be approximately constant, so none of that "15A max = 15A at all times" shit we get every week.

>>1400417
you will need a way of getting a sub woofer signal to the new amp. The best way to do this is with rca cables from the sub output on the receiver, but most stock units don't have that. The sub amp might have a high level input which can connect in parallel with the front speakers. Otherwise you will need a high level (speaker wire) to line level (rca cables) converter connected to a low pass filter/crossover which is connected to the amp. you need a crossover so the sub woofer only gets the bass notes

>>1401704
>Reverse would be nice but not at all necessary.


"This circuit is based off of a Driving Bigger Loads circuit in one of my books. That circuit only used one MOSFET and a diode. It is meant for a micro-controller to control a motors speed. To be able to go in reverse I just added a DPDT relay and another MOSFET, diode pair to control the polarity switch. I hope you enjoy this instructable."

Like that other poster said, the relay is to reverse. You should look at his Driving Bigger Loads for a simpler design, which might need different fets but should be simpler to adapt.

>>1401708
>in one of my books

OOPS. I read it too fast and thought he was referring to another blog.

>>1401705
>Coil is only needed to be able to reverse the motor.
I see, thank you very much for that information :)

>>1401705
>Will you have a current shunt on the line in to monitor current and a divider to measure the power used by the motor?

Would this be for programming the power restraint to avoid pushing the motor too hard?
My strategy for this was going to be using some testing and trial and error to work out what PWM ranges I can out put to the controller and find out at what point the controller outputs 36V@7A and I'd prevent the micro controller ever outputting that much.
Is my idea not going to work? Now I've typed it out, im questioning my understanding on how to measure the power output from the controller.

>>1401705
>The motor's resistance will be approximately constant, so none of that "15A max = 15A at all times" shit we get every week.
Could you explain what you mean by this?

>>1401708
>"This circuit is based off of a Driving Bigger Loads circuit in one of my books. That circuit only used one MOSFET and a diode. It is meant for a micro-controller to control a motors speed. To be able to go in reverse I just added a DPDT relay and another MOSFET, diode pair to control the polarity switch. I hope you enjoy this instructable."
So this is one of those times I could have found my answer on my own. I apologise for coming here when the answer was so clearly there all along aha. Regardless, thanks :)

>>1401708
>>1401709
I can't actually find the guy's book. Any idea where to find it? If not I'll just message the guy but his last activity was 11 months ago so idk about that.

>>1401464
Yeah I cooked enough of the oscilliscope that it wasn't worth dealing with
The half assed broken board was to fix another one enough to work so I couldn't finish what I was working on. Isn't 100% working so one those connections probably isn't good, makes an interesting color separation though so maybe I'll use it for audio

>tfw dumpster diving is not allowed in your country
God knows how many washing machines are out there rusting away with their motors still inside...

>>1401717

it's probably this circuit but maybe a different fet; there's some discussion here:

https://electronics.stackexchange.com/questions/179084/driving-dc-motor-using-a-single-mosfet-why-does-the-motor-spin-without-applying?rq=1

that site came up from a google search "arduino motor driver one mosfet"; if you search for "arduino motor driver" you get lots of L298 solutions which are for bidirectional drivers

>>1401722
third world shithole confirmed
I hope customs doesn't fuck you hard for chinkshit

>>1401712
current draw is also dependent on mechanical load. stall current is much higher than no-load current. you will probably want several milliohms (like 20 or 30) on the way to the negative rail so you get a voltage across it proportional to the current passing through the load (and it). negative rail bc that's the one your ADC is probably using as a reference, and size the resistor appropriately
if you're feeling really fancy you could use comparators that are internal to some micros to continuously measure against a voltage reference, possibly adjustable by trim pot, and trigger an interrupt to put the whole system immediately into a fault stop condition when current exceeds some value

>>1401737
>I hope customs doesn't fuck you hard for chinkshit
Last time I ordered 48€ worth of stuff from the US I had to pay 30$ of import tax.
Fucking Europe man, it's like they're trying their hardest to fuck you over.

>>1401736
Thank you very much, this appears to be exactly what I'm looking for :)


My 36V SLA battery (3*12V cells) outputs approximately 38.1V at full charge, and I've yet to see its voltage at a completely discharged state. If i push this changing voltage (with an estimate range between ~34V to 38V) through the MOSFET configuration attached, will I run into big problems?

I'm looking at using an LM317 adjustable voltage regulator to keep the input voltage at 36V. However, will this regulator only work when the battery's voltage is above 36V? Since i'm expecting it to drop below 36V, will the LM317 still regulate the voltage or what should I use instead?

>>1400396
There are eight 1.2V cells inside the battery. But it seems like the four on top are connected to the four on the bottom, maybe only by the plastic wrapping. Also there are little metal clips on the top and bottom that seem to be fastened to the batteries. I have no idea how to remove those temporarily.

>>1401736
>There are too many people showing us cartoons in their questions instead of schematics already. Stop making this worse by giving the impression these cartoons are acceptable here. They are not.
I like this guy already

>>1401789
>an LM317
>for multiple amps
I seriously hope you guys don't do this
>36V SLA
>38.1V at full charge
suspicious. you may have some bad cells?
anyway don't worry about voltage regulation, since your load runs on current and you have a feedback loop that senses current. voltage will find its own level as with any inductor, and the important part is feeding the motor the correct current to keep its load turning steadily. just derate (design your circuit for somewhat above the intended operating condition, say, 50V instead of ~38V, or 20A instead of 15A) for maximum reliability
but seriously, you gotta start reading some MOSFET datasheets and familiarizing yourself with their parameters

>>1401789
>I'm looking at using an LM317 adjustable voltage regulator to keep the input voltage at 36V.

I'm pretty sure that that regulator needs a volt or two above what you want, meaning if you want 36 at the output, you need at least 37 or 38 at the input. Also, without additional circuitry, it's only capable of about 2 amps. But I'm not sure you need to waste any effort on regulating the voltage since you said you are designing a a "motor speed controller". If that's the case, you will sense the speed and adjust the PWM to get the proper speed, which effectively regulates the voltage and current in the motor. And if the mosfet is rated for your maximum voltage it won't matter if the voltage is lower if the batteries can supply the current needed to drive the motor.

>>1401791
You neither can nor do you need to remove them.

>>1401812
>you have a feedback loop that senses current.

He originally said he was building a motor speed controller, and I wonder if we've steered him off the path. He needs to tell us what exactly he's going to sense and what exactly he wants to control. Controlling the current is fine if he has components that can be burned up, but if he's actually controlling speed it seems that he can analytically determine that every component can handle the maximum current that fully charged batteries can put into a stalled motor, and add current sensing/protection if he needs it.

>>1401812
>you have a feedback loop that senses current

Do I? Where? lol

>>1401838
>He originally said he was building a motor speed controller,
I (believe) I still am.


>He needs to tell us what exactly he's going to sense and what exactly he wants to control.

Sense? From what i've described, I'm not sensing anything.
In terms of control, I want to control the power output of my 250W motor.
Is my motor controller supposed to vary the voltage or the current to the motor?
Note: I still intend on using the configuration I posted here:
>>1401789

>>1401822
>If that's the case, you will sense the speed and adjust the PWM to get the proper speed, which effectively regulates the voltage and current in the motor. And if the mosfet is rated for your maximum voltage it won't matter if the voltage is lower if the batteries can supply the current needed to drive the motor.

I see, so if i understand you correctly then I don't need to worry about using a regulator circuit to stabilise the voltage supplied from the battery?
Please note that sensing speed is not entirely necessary, only controlling the power supplied to the motor is necessary.

>>1401860
>Where?
here, or at least you should be

>>1401860
>In terms of control, I want to control the power output of my 250W motor.

why? what does that mean to you? motor output power is speed x torque, and while that might be something you want to control (power, regardless of speed or torque), how can you control something if you don't sense anything?

to control motor output power you need to sense speed and torque and you cannot do that in a practical sense, not at your level in any case.

go back to your assignment, or your motivation, and figure out something realistic. controlling speed requires a speed sensor, which could be a hall effect switch and two small magnets on the shaft (two for balance). or some sort of optical sensor.

>>1401873

while I agree that lots of motor controllers sense the current and either regulate it for some purpose or simply keep it from exceeding a preset level, why do you say that he needs to sense it? The issue as I see it is that he doesn't seem to know (or be able to articulate) what his object is.

>>1401880
>why? what does that mean to you? motor output power is speed x torque, and while that might be something you want to control (power, regardless of speed or torque), how can you control something if you don't sense anything?

I think the confusion here is that you're assuming i'm building an automatic system. This is not the case.What I'm trying to build is a system wherein a throttle sends a voltage signal to the microcontroller, that signal is processed and a PWM signal is then output from the microcontroller to the motor controller. The PWM signal is used to control the power output by the motor (or rather, to control the power output of the motor controller which in turn controls the power output of the motor itself).
The 'sensing' would be done by the user.

>>1401873
That's not quite the configuration i'm using.
I've attached what I'm using. Which would be better?

>>1401885
>motor controller

ah. I'm using "control" in the system sense, where some signal is used as feedback. what you are planning to do looks fine.

You probably already know this, since you haven't asked, but the throttle can be a potentiometer between V+ (check your adc spec to see if that's 3.3, 5, or something else if you are providing AREF...) and ground, and the wiper connected to an analog pin that reads the voltage and sets the pwm from 0 to 100 percent, or perhaps a smaller range depending on your needs.

>>1401892
if you want to sense the current drawn by the motor and don't mind frying the transistor when t he motor turns off you need to use >>1401873
if you don't care about current measuring >>1401892 is basically the same

>>1401920
Is there any way I could do both? As in, use my method and modify it to measure current as well?

>>1401929
work out the max current the motor will draw
use the max voltage the adc can handle, 5v?
R=V/I gives you the MAX resistance to use.
check power rating too P=I*V.
connect resistor between GND and FET Source. connect FET source to adc input.
calculate current through motor as:
((adc value / max adc value) * 5) / R

Using truestudio (eclipse based IDE), I made a header file to include in main.c, but if I include the same header in the interrupt request file (literally in the same folder as main.c), the compiler won't include it. Any ideas as to why this would be?

I've got some 12v 1.5W solar panels I wanted to use to charge a USB battery bank that's rated at 1A input, using a buck converter to get the panels' voltage down to 5v.
The battery bank charges properly when there's no load attached, but doesn't when I plug something into the battery's USB port. Do I need 8 of these fuckers to get the battery to charge reliably under load?

>>1398409
bump for same feeling and deepening curiosity

>>1398409
>sub
You have to go back

>>1401966
some power-path manager ICs share the switching inductor between charging and discharging, and therefore can't do both at the same time. some battery banks just refuse to because fuck you I'm not a UPS and you're not my real dad
what you need is something (conceptually) other than a USB power bank. I've successfully used the bq24195 IC followed by a boost converter to serve as a UPS-like battery backup system. if you'd rather not have a microcontroller around to set up the bq24195 away from its conservative default settings, there's also the MCP73871 that is truly standalone. in either case, enjoy your 0.5mm QFNs with exposed pad

Say I have two separate parts of a board

I need to separate them in a way that I can place something like jumpers if I want the signals to pass through from one part to the other OR disconnect the jumpers and connect an external cable if I need it that way

What would you recommend for this situation?

A header row would be the obvious answer but it's a bit too wide, I have about 7 such signals and I'm width-restricted, up to like 1.3 cm in width total

>>1401966
gonna depend on how much amps are you sucking from it clearly you are taking more than 1 amp so theres not gonna be enough "pixies" to charges the battery and at the same time power other things , you need more amps according to how much you are taking.

>>1402035
I'm running an ESP8266 with a couple sensors attached, so I'm pretty sure I'm not pulling an amp - but fair enough, I'll keep adding solar panels until it just werks

>>1402032
I'll just hot glue that bitch in

Babby's first surface-mount soldering project is a success
I mean, the board is a mess, the headers are crooked, I completely destroyed a trace, and I fucked up soldering the IC so badly the first time that I wound up cutting it off the board and buying a new one
But I can flash the "blink" program from the Arduino IDE and the goddamn LED blinks, so I'm calling it a win

>>1402060
did you solder that with a potato
ah fuggit, it works

>>1402032
In BigClive's latest power-bank teardown he mentions people slapping a standard charge controller circuit in a normal power bank that allows it to charge and discharge simultaneously.

>>1401791

if your plan is to separate the cells, charge each one, then re-assemble them, then this is the worst plan ever. DONT DO IT. if you cant make your own charger, (which is really simple but you need an ammeter and several small resistors) then your next best bet is a 9.6V charger used for RC cars. just connect the two using alligator clips or paper clips, or whatever you got.

>>1402106
I have every resistor value under the sun and by ammeter you mean a multimeter? I got one of those. I'll happily try to create a charger. But how do I go about it?

>>1402034
Terminal blocks?

>>1402034
what about a 2 * whatever header.
use a single row ribbon cable to connect the left row to right row, or use a block of jumpers like pic related.
total width 0.2"

>>1402265
oops sorry no pic.
see https://www.digikey.com/product-detail/en/samtec-inc/MNT-106-BK-G/SAM10621-ND/2685313

I want to use transistors insetad of relays, what kind of transistors should i ise to work with computer tier voltages? (so 1 - 10 volts and 12 - 24V, while the trigger current will be 3V from gpios)

>>1402274
Does the output have to be a transistor?

The main benefit of transistor is longer life time and faster switch time, because there is no physical switching.

A none issue unless you really need it.

>>1402276
Well i basically need it for very simple hobby projects.

For example, I want raspberry pi to be able to power some device that needs 24V.
Pins on the RPI can only supply 3V at maximum.
So i want to use transistor for this, where i snip the 24V wire going to the device and put a transistor on it, and then connect the thrird leg to the RPI 3V pin, and this will allow me to simply turn the 24V on and off with the pin
I know i could use a relay, but I am lerning electronics and I really love the concept of these magical transistor relays with no moving aprts
so can you recommend a transistor name capable of this?
I looked at ali and they have so, so many of them it's very confusing

>>1402282
But you can use the one i posted for that.

http://denkovi.com/relay-module-5v-8-channels-for-raspberry-pi-arduino-pic-avr

>>1402284
That is a whole board tho, i am talking about those tiny black things like in pic related which i an use in my bread board

>>1402219
>But how do I go about it?

already told you above >>1400647

get a transformer above 10Vdc, preferably around 12-15V, and put a current limiting power resistor in series with it. try various resistors while using ammeter to measure a current that's safe when the battery is all run-down. say 0.1 to 0.25A.

>>1402286
Ahh.

Better of with a mosfet then, if you plan to switch the load directly.

Watch this video.

https://www.youtube.com/watch?v=GrvvkYTW_0k&t

>>1402286

just choose any transistor that exceeds your max. voltage, and max. current requirements.

here are 4 parts that will handle 90% of all jobs.
2N3904 BJT, NPN
2N3906 BJT, PNP
TIP120 BJT, NPN (Power),
FQP30N06L MOSFET, N-Channel

>>1402293
Thanks.
One more thing though, i googled the specs for some of the transistors sold and this is pretty confusing,
where do i find out the important stuff like maximum power this can handle, and minimum trigger voltage?
https://www.onsemi.com/pub/Collateral/P2N2222A-D.PDF

>>1402300

in your example,
- VCEO is 40V so that's yr max voltage.
- IC is 600mA so that's the max current
- PD is 625mW so that's the max power it'll dissipate
- VBE(sat) is 0.6V which is what you must present at the base to turn it on
- VCE(sat) is 0.3V which is how much voltage you'll lose across the transistor

>>1402302
so if it says that VBE(sat) max is 1.2V and i try to use it with rpi 3V pin it will burn right?

>>1402304
>so if it says that VBE(sat) max is 1.2V and i try to use it with rpi 3V pin it will burn right?

that's why he showed you the base resistor a moment ago.

>>1402304

nope, the series resistor in the base (diagram in >>1402293) will eat up the excess. that's its only job.

>>1402305
>>1402306
I know i am asking stupid questions, i am trying to grasp all of this
that is pretty cool, i had no idea they had built in resistors like that so you can use bigger range of voltage to switch them on and off

>>1402308
>built in resistors

sigh.

>>1402308

the resistor isnt built-in. this way you can change it so suit a certain purpose. for example, with a 3V input, you'd probably wanna use 1K instead of 10K to make sure it turns on hard and fast.

>>1402310
Ah, I see, so if the voltage is have is too high i can use the ohms law to calculate the resistor size and then use it to control it safely anyway

>>1402311

yeah, that's a close-enough understanding of it to get by.

>>1402315
The learning curve for this stuff is pretty high for me since i was never good with this stuff.
I ordered several hundred leds, resistors, transistors etc so I can just keep destroying them until it clicks.

>>1402075
>appeal to internet authority
that's a shitty hack and not particularly safe or reliable

>>1402034
there are headers in smaller than 2.54mm pitches, readily available. consider right-angle versions so that you can just push the boards together, if that works for your mechanical design

>>1402282
the next question is, do you want to switch those loads on and off from the high side, or the low side? the solutions so far are all low-side switching, which may or may not work for something that connects to other things e.g. a speaker amp, or loads that are connected to ground by other means e.g. automotive stuff. so you use the low-side switch to control a high-side switch as in Pic related

>>1402322
The actual application I will want to use it for is to turn on an off a metal heater plate which is powered by 24V that (those I will be getting from an old PC PSU)
In care i won't be able to make it work with a tran, or a mofat, i will just use a regular old mechanical relay that is guaranteed to work

Are these interconnected in any way? Doesn't look like it from the drawing.

>>1402334

I don't think they are. Veroboard aka stripboard has all the holes in a row connected, and you cut them where needed, but otherwise they can make connections easy, like having a ground row, +5 row, and just using them to make any connections. Like adding a header beside an IC; with strip board it is trivial.

But that kind you show has limited use in my opinion. The plated thru holes make it easy to solidly mount a part, but as for connecting, you either need a huge solder blob or a wire of some sort.

I guess electronics general is good place to ask. Where can i find updated versions of software for drawing schematics? I was using see electrical but i have severely outdated version

>>1402349

https://thepiratebay.org/search/altium/0/99/0

if altium is more man-meat than you can handle, search for the following titles while on the PirateBay : https://en.wikipedia.org/wiki/Comparison_of_EDA_software

If i want more current to go trough a transistor, does that mean i need more voltage to punch it trough at the base?
I have this simulation program, and i just don't get it.
It's PNP.
I thought that if you need say 5V at the base for it to open, then it doesn't matter what you pour trough it, if 2V goes in, 2V goes out, if 500V goes in 500V goes out (same for current). But when i run the simulation i get completely different values.
40V in, and 30V at the base (i cranked it up to make sure it fully opens), nets me 17.3V out.
What the shit is going on here? Is the simulation app broken?

>>1402360
I mean't NPN not PNP

>>1402360
>does that mean i need more voltage to punch it trough at the base?
More current to be precise. An NPN or PNP transistor like that has an output current that is proportional to the base current, so decreasing your resistor value and/or increasing the voltage at it will mean more base current and therefore more current through your load. Ensure that your transistor is operating in common-emitter mode, else you'll be getting negative feedback by virtue of the base-emitter voltage floating.

Pic of circuit

>>1402364

>>1402366
What produces these absurdist paintings?

>>1402371
It is called Every circuit. I am open to suggestions, but it has to be drag and drop and have simulator, where you can run the circuit and see the voltages and currents and things like if you managed to fry a component, since it is great for learning

>>1402372

your simulator is just wrong. if you put 30V at the base in that kind of circuit, you'll get approx 27V at the emitter. maybe replace the LED as a load by a resistor which might bring the simulator back to sanity.

>>1402287
Ok thank you. Sorry for all the questions. As a beginner this is all very intimidating. But I think I understood you perfectly there.

>>1402372

your simulator is just wrong. if you put 30V at the base in that kind of circuit, you'll get approx 29V at the emitter. replace the LED load by a resistor which might bring the simulator back to sanity.

>>1402372
LTSpice >> anything else that doesn't have spice in its name. Has transient and AC analysis, plus user definable parts if you can figure out how to get that to work. 555 timers also.

You're trying to measure the behaviour of a current-controlled current amplifier with voltage sources and nothing to limit current at all. In the real world, transistors are used either in the linear region, as an amplifier, or in saturation, as a switch. Both of these require a base resistor and a load resistance, neither of which you have. Pic related shows an expected relationship between base current and collector current (base current has been multiplied to be more visible), linear region on the left and saturation region on the right.

>>1402353
I really hope that isn't what anon was looking for. Bless him if he was using SEE electrical for that kind of stuff.
>>1402349
There is a student version of it, if that works for you?

>>1402325
then the low side switches presented by others will be fine, and a MOSFET would be much better and safer for the application than a mechanical relay (won't weld closed)

>>1402377
his circuit is just wrong. the load rightly belongs on the collector of a pnp like he was trying to draw the first time

>>1402401
>his circuit is just wrong

nah, the circuit isnt wrong, it's just different than what you're used to. it's called a common-collector configuration, and works essentially as a current multiplier, as opposed to an on-off switch like the common-emitter.

>>1402401
>mechanical relay (won't weld closed)
I will be working with battery tier voltages (12 - 24V) and the relay is rated for up to 250V outlet, so i doubt that could happen

Hey anons.
I don't really know anything about electronics but I have this problem.
This one computer parts store sells there computer fan adapters. You connect the fan into it and then the adapted into the motherboard and it reduces the fan's speed by say 50% or 70% depending on which adapter you use.
Those shitholes charge ten bucks for it.
I examined the adapter and it is literally just female fan connector ---- 3 wires ---- male fan connector
Except, one of the wires has this kind of bump on it, it looks like a resistor.
Can you tell me how could i replicate this effect? What kind of resistor do i need to place on which wire of a 120mm PC fan to slow it down by 30%, 50% and 80% ?

>>1402412
> it looks like a resistor.
got a pic?
It's probably in the 5-60 ohm range.
Get a handful of these e12 values and play around.
It could also be a diode or even zener diode, but try resistors first.

>>1402412
PC fan speed is usually controlled by voltage where 12V is full blast, they probably just have a resistive divider in the cable

>>1402420
I can't look at it since it is encased in some black rubber seal thing
But I once tried connecting a second fan to it, so I could use one of them to reduce speeds of two fans, and it got so hot it started smoking.

>>1402428
Yep sounds about right.
The device works by wasting electrical power as heat.

>>1402453
you can't waste energy anon, only convert it


>the more you know

>>1402487
I said waste, not destroy. You know perfectly well what is meant by waste heat, it is commonly used term with a well-understood meaning.
What is the point of your autistic pedantic faggotry?

If i have battery connected to a light bulb and put a capacitator around the light bulb, does that mean bigger capacitator = it stays on longer after i remove the battery? Can i use any capacitator?

hello, I already have some experience with 8051 and assembly in the arduino chip, but I don`t know anything about other MCs. I want to make a datalogger that saves things in a SD card. I`m thinking of getting a chinkshit PICkit 3, is it worth it? what PIC would be good for the job? Or should I look into MSP?

>>1402502
>does that mean bigger capacitator = it stays on longer after i remove
higher capacitance = stays on longer
>Can i use any capacitator?
yes as long as the capacitor's voltage rating is higher than the battery's

Hey /diy/,
I need a scope. Nothing too fancy, but I want to diagnose output signals from various electronics.
Anything recommended that's not expensive as fuck?

>>1402522
But how about the discharge rate? How does the capacitor know how fast to discharge the current (how many Amps to pour into the bulb) ?

>>1402520
Depends on PIC, but generally speaking: if there's one thing which particularly sucks in PICs, then it's the memory handling. Not that you can't use them, but of all the processor families you mentioned, they're the most cumbersome choice for pushing blocks of data around. Using C hides most of it, though.

>>1402530
How does your lamp know how many amps to draw?
Point being, your lamp determines the current draw and that current draw determines how quickly the capacitor's voltage drops. You need pretty big capacitor to keep the lamp on even for a second.
Assuming constant current load (often bullshit assumption), the capacitor voltage drops I/C volts per second. That is, 1000µF capacitor and 1mA load = 1V/s discharge rate.

>>1402551
So what PICs would be good for the job? I manage in C and full autistic assembly. I don`t know anything about PICs or other MCs. Like which one to use for that or this thing.

>>1402577
And should I go PICs or MSP and buy a launchpad?

I bought a Raspberry Pi 3 on a whim while at Frys but I can't think of anything to do with it that I couldn't do with a $5 or $10 model
should I just return it and buy a Zero online?

I'm going to be driving back there anyway to return something else (a starter electronics kit that was $30 off but I found out I was still jewed out of another $40 after comparing part prices on alixpress)

Anyone have a recommendation for what to buy if one wanted to get into arduino? I just got an uno from one of my classes but don't have a breadboard or anything, dunno if I should buy one of the starter kits on ebay or just purchase a bunch of shit I want to mess with.

I'm going to school for electronics so I have kind of a head start, just want experience coding this shit + they look fun to make projects with, anyone have reccomendations on what I should start off with buying or maybe some good beginners projects?

>>1402611
maybe some computing-intensive task? Something that needs a lot of ports?
>aliexpress
some chinese "knockoffs" are cheaper also because they use cheaper components, so watch out for that.

>>1402612
maybe one of those component starter packs but to be honest they're not even that great. Yes they do give you a bunch of different things but so many of them have like 50000 capacitors and yet you'll never use them because they have all the values except the ones you will need for your project. But if you really have nothing at all then I guess it would be a start... Otherwise I saw a while ago that there are these assortments of arduino sensors? Dunno about the quality though.

>good beginner projects
Look at the OP or if someone is kind enough to post the image (also seen it posted within the last couple of threads i think)

>>1402612
https://www.youtube.com/watch?v=oJziPSawlF8

>>1402589
toss a coin, it basically doesn't matter for simple data recording. If you want someone to pick for you, go MSP and get out of your comfort zone. Read the launchpad description and its processor's datasheet summary to be sure it'll do what you want.
8bit PICs are dinosaurs but very mature and feature packed and have higher current gpios than MSP. And usually wider operating voltage range.
MSPs are 16bit and do more per clock cycle.

>>1402611
Not really super fun project wise but retropi is pretty fun if you're into vidya. can run N64 and PS1 games surprisingly well.

Is going to school to become an electrician a wise idea?

>>1402619
Yeah I might just end up buying the components I want and starting with some simple projects, those kits look neat but there's a bunch of shit in there I probably won't end up using.

>>1402622
Doesn't really help me decide but that was a really cool watch

>>1402645

- are you short, stout, and not too bright?
- do you love the idea of charging a 300% markup on parts?
- when you bend over, does your ass-crack show?
- do you think it's okay to replace perfectly functional parts?
- can you withstand electrical shocks below 480V and survive?
- do you like to get dirty and greasy?

if you answered yes to all these questions, then an electrician's job is right for you.

>>1396689
>tfw have been storing two small neodymium magnets in the same cabinet where I organize all my electronic components

I am paranoid know, did I fucked up? They where "isolated" in their own compartment but still some cm away from my, transistors, diodes, chips, wifi modules, micro-controller, etc.

Can you explain the difference between these two?

>>1402792
Anyone?
I basically need to know what effects ahs the placement of the power source fo the LED

>>1402792
The right one has constant tension on the Anode

>>1402855
The left one*

>>1402855
If i measured the current going through the led, would it he identical in both cases?

>>1402772
Magnets won't hurt those at all, they're just doped silicon and copper and gold. If an IC has a magnetometer inside it it might make a difference, however.

>>1402792
There's no point in running a transistor in common-collector mode if you're not limiting the base current, and even if you are you're not going to want to rely on the hFE of a transistor remaining stable.

By running your collector to ground in the left image you're not using it like a transistor at all but rather you'd be shorting the voltage source to ground through the b-c "diode", and passing way too much current through the LED if the transistor can handle those conditions (which it can't). On the right image you'd have a somewhat limited current through the LED (provided the base voltage is lower than the collector voltage), but this current is highly dependent on the behaviour of the LED.

Use a common-emitter arrangement with a current-limiting resistor in series with the LED instead of common-collector, provided you just want to switch the LED on and off.

>>1402881
I am desperately attempting to design the circuit in a way where the transistor acts as a simple "mechanical" relay. But no matter how I design it I always get wonky behavior
Can you provide as simple as possible circuit example to achieve this?
The trigger current for the tran. i will use will be 3V and the "relay" current will be 24V

>>1402884
R1 is there for current limiting to the LED (see LED resistor calculations) and R2 is to limit current such that the transistor is still saturated but too much power isn't being wasted through the base. A standard transistor will have a minimum application factor of about 50, so we take our maximum collector current (in this case 20mA) and divide it by 50, to get 400µA. Dividing 3.3V by this current gives us the maximum resistor value of about 8kΩ, so we round down to 5k to ensure that the transistor is turned on hard enough.

But using an N-channel MOSFET would be a better idea in general.

>>1402888
So my example wasn't working because i was missing the resistors? I wasn't using them on purpose to keep it as simple as possible (in this case i didn't cared about burning the LED) are they the sole reason why it wasn't working?

>>1402895
You were using a common-collector arrangement (collector to a voltage rail), while I used a common-emitter arrangement (emitter to a voltage rail) so the position of the load was at the collector instead of the emitter. It means that the base-emitter current is independent of the load resistance and is much easier to get into saturation (which you want for a switch) than a common-collector arrangement.

But in general if there's a path from one rail to another without going through a resistor you've done something wrong.

>>1402864
>identical in both cases?
Yes, zero, because of the floating bases.

>>1402908
>You were using a common-collector arrangement (collector to a voltage rail), while I used a common-emitter arrangement
Oh I see. I had no idea that it matters on which side of the transistor the components are

>>1402895

you were given two perfectly good designs in >>1402293, one with a BJT, and another with a FET but instead of using those tried-and-true designs, you're inventing your own Frankenstein arrangements that dont make any sense. i'm thinking you may be either a troll or a complete fool.

>>1402920
There is a difference between just copy pasting something and trying stuff out to actually understand why it works.
So that way when i assume how something works and then change it and it doesn't I know the assumption was wrong.
I am sorry if that makes you butt crack sore.

>>1402921
>I am sorry if that makes you butt crack sore.

ah, i was wondering why my butt was itchy. it was you in there all the time.

Will it blow up?

AGND and GND is connected internally.

>>1402921
>>1402921
M8, It`s a circuit, not a girl. Nothing is stopping you from understanding how it works. Every component there has very well defined behaviour. You can copy and paste and then make the analysis.

>>1402974
Unlike say programming ,where you can fail over and over again with no consequences, electronics are different. A single mistake can cost you hundreds of dollars in destroyed parts, not to mention your life, if you finger the wrong amper-hole.

>>1402979
>Unlike say programming, where you can fail over and over again with no consequences

"The most devastating computer virus to date is MyDoom, which caused over $38 billion in damages"

you've chosen a bad example; programming errors have cost society hundreds of billions.

>>1403027
That is not even a strawman. That virus was created on purpose and it took a massive skill and knowledge. You can't create a computer virus by accident.
That is like saying that accidentally reversing polarity on a battery is the same thing as taking your phone and smashing it with a rock.
Also no matter how complicated the code you can test it safely on your own computer and have it fail without damaging anything. Only an idiot would run untested coded in a prod setting
This is untrue for circuits where you will destroy all your expensive test parts and test devices.
You are comparing apples and oranges in terms of damage costs.

>>1403029

god, you're dumb. viruses only work if there are bugs that allow them to do so. every virus exploits one or more bugs to propagate. so, yes, program errors have consequences. moreso than any other errors except war.

>>1402979
Programming and electronics have little random things. They are very precise areas if you do things right. Transistor behavior isn`t anything new. That guy was just stupid.

NEW THREAD
>>1403081
>>1403081
>>1403081
NEW THREAD

>>1403046
Are you a moron? Programming had zero consequences when it comes to practice. I seriously hope you are just trolling

>>1399693
>>1399700
>>1399694
thanks so much guys. so linear does has its place and so does switching. i should learn both.

the capacitor should use diode + resistor like a large ohm resistor but one that will drain the capacitor in short period. basically a flyback config except this would slow down volt + current at safe level

seriously, i appreciate all the replies.

p.s i know caps are dangerous/lethal and i treat them as charged until i use my DMM to check voltage.