/diy/ - Do It Yourself

so the arduino atmega has a device max output current of 200mA and a pin max off 50mA (or something like that)

you guys think i could pull 200mA through one pin for a few ns without causing any weird internal voltage drop effects? i'd like to turn on a fet without any driver circuitry besides a resistor and 50mA leaves it switching pretty slow.

>>1102003
Pin output current isn't ' how much current is this pin capable of supplying' it's actually "how much current can I pull through this pin before the insides melt and parts of it stop working in strange and annoying ways"

You can pull current over the stated maximum and it will probably work but you will definitely cause long term problems that are difficult to diagnose

>>1102007
consider though that the peak is only going to occur for 10ns at most. that's a pretty small i^2s that then has >1000ns to equalize before the next edge. my concern isn't with vaporization, since i'd imagine if the pin drivers were fully characterized and had a SOA in the datasheet i'd still be within it.

don't get me wrong, you could be right. i'm just more concerned with the possibility of voltage drops caused by the current spike resulting in some fucked up logic. i know fets reasonably well but i don't know much about digital ic topologies.

>>1102010
Got a scope? You could always check. I'm curious, too. I wouldn't worry about anything frying, either.

if i take four 2.7v 400f supercaps and hook them up like so, i should get 5.4v 400f at the load right?

>>1103181
No. Caps in series reduces the net capacitance like this Cnet = (1/C1+ 1/C2), so you'd only get 200F.

Putting caps in series to increase voltage tolerance will usually work, but things can go wrong. If you put 5.4v on them it won't be distributed evenly if one cap has a higher leakage current so one will go over the 2.7v limit.

Also you should be using a capital F for Farads.

>>1103181

in theory, yes, but for practical purposes, you'd add an equal resistor across each cap to make sure the voltage equalizes in the series caps. size of resistors is a tradeoff between effectiveness and discharge rate. the right value will be somewhere between 1 ohm and 1 million ohms, i'm sure of that.

>>1103185

nah, dude. 200F in each branch, so 400F total.

What is this thing on a circuit diagram?

Why are there tildes, minuses, and plusses?

>>1103250
If you're not trolling, it's a diode bridge.

>>1103250
to expand, it's just 4 diodes in a convenient arrangement exactly like it shows. if you wire an AC source to the tildes, you'll get a rectified sine wave on the +/- which can then be filtered into dc.

>>1103257
>>1103271
Thanks, anon.

I'm not trolling. I'm a complete noob.

just bought a variac from circuit specialists and its pretty good, however the voltmeter on the face and the printed one on top are inaccurate as it is meant for 110v but my house has 125v. at 110v its supposed to scale from 0-130v but on 125v it goes up to 150v. of course this is fine, but the needle shows 150v as 130v, and scales the whole range incorrectly as such. i want to either calibrate the voltmeter it comes with (its just a cheapo chinese one) or just replace it preferably with a simple digital one. anyone know how i would fix the scaling or of a replacement that would work with these dimensions?

>>1102003
The 200mA is the total it can supply across all pins combined.

>>1103662
Replace the dial markings. Or cover them with your own. Or glue proper ones on the outside.

Also those usually have a calibration potentiometer somewhere.

>>1103695
I don't think I could do the dial ring properly with getting the spacing accurate and such, dont even think they sell a 150v one. I popped open the front panel and all it was for the gauge was a cheap plastic gauge iirc in series with the output plugs with no sort of adjustment.
This is the exact one I got https://www.circuitspecialists.com/variac-tdgc2-2.html

>>1103877
A voltmeter is a voltmeter is a voltmeter. It's not "calibrated" for one voltage specifically, it doesn't know or care what the input voltage is if its 110 or 125 or 240 or 20, if its showing the wrong voltage then it's just shit. Look for a tiny screw to turn to calibrate it at the axis of the needle.
You can buy panel meters for a few units of currency to replace it with, just make sure the input voltage is within range.
A digital panel meter might need a power supply e.g. Separate 12v dc input to power the lights.

>>1102010
It's not going to melt or vaporize but it will break down the insulation inside

>>1103886
This is true. I'll take a second look for that screw when I get home

>>1103250
It's an AC/DC rectifier boss man

>>1103886
Just checked and I can't adjust it. On the front it looks like there might be an adjustor but it's behind the plastic and it seems plastic welded together rather than with clips.

>>1103955

>>1103886
They do care if it's AC or DC and have upper limits of the voltage they can handle without damage. Stick AC on a DC galvo voltmeter and watch the needle shake back and forth.

>>1103956

- the calibration screw is that red dot at the center bottom. looks like it's been painted over to avoid tampering.

- before you go adjusting anything, you have to make sure you're comparing it against a known good meter. if you're just guessing, then assume you're wrong and the meter is right.

- if the meter is reading too high, you can add a resistor (or pot) in series with it. if it's too low, you MAY be able to add a resistor in parallel. it depends whether the thing has an internal current-limiting resistance (in which case it wont work), or an external resistance (in which case it will work).

- if you're wondering why the meter goes to 150, it's coz variacs actually step up the voltage as a consequence of how they operate.

>>1103956

- the calibration screw is that red dot at the center bottom. looks like it's been painted over to avoid tampering. it's not really for calibration, it's for zero-ing but it can used to to adjust the readings up or down a bit.

- before you go adjusting anything, you have to make sure you're comparing it against a known good meter. if you're just guessing, then assume you're wrong and the meter is right.

- if the meter is reading too high, you can add a resistor (or pot) in series with it. if it's too low, you MAY be able to add a resistor in parallel. it depends whether the thing has an internal current-limiting resistance (in which case it wont work), or an external resistance (in which case it will work).

- if you're wondering why the meter goes to 150, it's coz variacs actually step up the voltage as a consequence of how they operate.

>>1104053
The screw is behind the front plastic and there's no way to open it.
I'm using my true rms multimeter for actual voltage.
I don't know what it has but it's obviously Chinese crap so whatever standard that comes with.
I know Variac's can step up voltage at the end of their range.

>>1102010
compute the average current draw based on your duty cycle, frequency, rise time, etc. if that value is greater than 50mA, you should use multiple pins to act on your load. if that value is greater than 200mA, then there is no hope for you. make sure you have a good amount of power supply bypass capacitance for the microcontroller.

either way, you're being a lazy cunt. use the microcontroller to act on a FET driver like a normal person. the architecture is standard for a reason.

>buy 2x8 lcd display
>hook up everything to a nano, pinouts and everything setup to either the datasheet or adafruits generic lcd wiring guide
>contrast is non-existent and never displays properly

i found some post online where a guy recommended to hook up a negative voltage line to the contrast pin (3) on the LCD screen. i put it in the LCD/arduino circuit, and put -4.5v on the contrast pin, and now lo and behold the contrast works, increasing the negative voltage (from -4 to -1) supplied lowers the contrast, and lowering it (-4 to -5) increases it

did i fuck something obvious up? i was hoping to use this LCD with just a standard 5/10/20k pot like i've done with 2x16s before, but i can't get this thing to display properly with them for the life of me. when P3 is hooked up to the wiper between GND/5v, the text on the screen is barely visible, and the changing the pot doesn't do anything at all, and i've tried different pots too, same effect

What's the best circuit simulator on Android? Doesn't have to be too complex, I'd like to fiddle with 555 circuits and the like.

What is the op image? It looks resistor-y

>>1104227
looks like a stepped variable resistor pot

hey /ohm/ idk what to do with my stuff left over from last semester
what are some cool things i could build with my parts along with my analog discovery 2
what i have:
>passives, assorted values
>some center tapped xformers
>assorted diodes
i dont have any opamps... where can i get some free opamp samples online?

So I'm an idiot and my knowledge of electronics is pretty much limited to hobby tinkering.

Anyway, I want to wind a power transformer, since I'm looking for a configuration that isn't all that common, can't find one locally and it'd cost an arm and a leg to import one into my country.

I've read a bit about the theory and following some equations I've reached the design parameters that i need for my PT. But i have a few questions about it.. :

- Given the calculated core area, i assume i can't go any lower to avoid saturating the core. But will there be a problem if i use a larger core area? There's a larger transformer that i might salvage the plates.

- The number of turns of each winding seems to be ultimately derived from the desired secondary current, i think. I've noticed that higher output current leads to less turns. I only need 0.075A at the output and the transformer I've calculated has over 1000 turns in the secondary winding, screw winding that by hand, lol. If i lower the number of turns of each winding accordingly, what's going to go wrong?

- Can i wind a primary, then a secondary, and then a separate "tertiary" winding for 2 outputs, or do i have to use only a single secondary with taps for 2 different voltage outputs?

>>1104240
>1
you can use a larger core. it'll allow you to use thicker gauge wires and will decrease both magnetic and copper losses.

>2
here's the order of determination:

Np (primary turns) = V/(4.4*f*Bmax*Acore)
Ip (primary current) = Pout/Vin_min
Np*Ip^2 ∝ required winding area, same for other windings

if you decrease Np you have to increase Acore or f (hello smps), or else Bmax (peak flux density) will increase. B saturation for iron/steel is 1.5 to 2T. approaching that limit is okay for mains freq designs because magnetic losses are low proportional to core size, but if your calculations don't account for parameter tolerances then you can get core saturation and then you'll be really fucking mad that you wound all those wires for nothing.

>3
there's no reason not to prefer a tertiary winding to a secondary tap unless you're limited in how many wires you can run to and from the core.

addendum: get an E core and a nice wholesome bobbin. toroids are for masochists. if it's not already obvious, buy magnet wire. if you're using standard insulation you're wasting a ton of winding area.

Alright senpaitachi, so I've been buying raw solar cells because that shit's cheap, and soldering strips everywhere applicable.
Is there any reason I can't just solder to one single point on the top/bottom of the cell?
Must I have constant connections running the length of the cell?
Any two point on the top (or bottom) have continuity, is it just that if I don't, drawing too much current through the traces on the cell will melt them, thus we solder to everything so there are lots of nice little points to pull current off the board?

>>1103250
It's a rectifier you silly willy.

Anyone here with a Uni-T UT216C clamp meter? I just got one for Christmas, and it all seems to work fairly well, except that it's reading itself for DC amperage. If I have the jaws closed, it reads -0.2A, and as I open the jaws, the reading goes up to around 0.18. I just wanted to check if it being this fiddly is normal, or if I got a dud. If I just know that I need to account for it in the readings, 's'all good, it'll still work.

>>1104308
Very helpful. Thanks a lot anon.

>>1104227
It's a stepped attenuator, its like a potentiometer but it switches between the different resistance values, rather than the carbon track and the wiper of a pot. they're on some high end amplifiers and audio equipment. pretty cool imo, they're really smooth to use and feel nice lmao

>>1104347

it works off of tiny magnetic fields, which you're creating by opening and closing the jaws. so, it's an artifact: an illusion, like freedom.

Hey guys, I'm trying to think of a useful DIY electronic project, and I am wondering if this idea is practical. I am trying to prevent my amplifier from damaging my speakers, if an aux cable is plugged into or removed from a device during use. Would it be possible to deal with this spike using a filter of some sort in series with the aux cable? I would prefer a solution that does not require modifying the amplifier itself. Any ideas you guys might have would be really helpful!

>>1103250
Speaking of rectifiers, I have a 12V 2A AC source I want to convert to DC, Do I just build that pattern with diodes and attach it, which diodes? Can I just buy packaged ones for a couple of cents somewhere? Will rectifying it introduce any noise or flicker?

>>1104404

if the amp uses DC-blocking caps on the inputs, which it likely does, then those caps will charge/discharged when something is plugged in/out. this creates a yuuuge input spike. one way around that is put a permanent resistor across the input in question. you wanna use the smallest value resistor that does not reduce the volume appreciably, something like 1K-10K.
since you're too chicken-shit to open up the amp, then you can make a short male-female extender cable, which incorporates the resistor in parallel with the input.

>>1104406
>which diodes?
any diodes rated above the expected voltages and currents that are gonna be used. simplest for you would be use something like the pic, from amazon. it's way overrated so it's likely to survive even handling by a noob. it contains all 4 diodes already so you cant screw up the polarities.

>Will rectifying it introduce any noise or flicker?
you're starting off with a ton of 60Hz noise, which will be converted to 120Hz noise by the diodes. to filter it out, add a big cap, like 3300uF at 16V across the DC outputs: plus-to-plus, minus-to-minus.

>>1104404

if the amp uses DC-blocking caps on the inputs, which it likely does, then those caps will charge/discharged when something is plugged in/out. this creates a yuuuge input spike. one way around that is put a permanent resistor across the input in question. you wanna use the smallest value resistor that does not reduce the volume appreciably, something like 1K-10K.
since you're too chicken-shit to open up the amp, then you can make a short male-female extender cable, which incorporates the resistor in parallel with the input.

>>1104406
>which diodes?
any diodes rated above the expected voltages and currents that are gonna be used. simplest for you would be use something like the pic, from amazon. it's way overrated so it's likely to survive even handling by a noob. it contains all 4 diodes already so you cant screw up the polarities.

>Will rectifying it introduce any noise or flicker?
you're starting off with a ton of 60Hz noise, which will be converted to 120Hz noise by the diodes. to filter it out, add a big cap, like 3300uF at 25V across the DC outputs: plus-to-plus, minus-to-minus.

If you short circuit a voltage divider circuit across the output terminals, does no current pass through r1? (the resistor which has the same voltage as the output voltage)

>>1104437

all the electrons flow thru the short, none thru the resistor, except one electron named Ahmed, coz he's the world's most non-conformist electron.

>>1104440
That makes sense I suppose. I guess there is a near infinitely larger draw for electrons through the short circuit than through the resistor, but there is only a finite number of electrons so they all go through the short circuit

>>1104437
Current would still flow in the resistor by Ohms law but since you short circuit the resister the voltage across the resister is 0 and then 0 current. But that is only in theory in real life you might get a few pA since the short circuit is not ideal

>>1104308
Where did you learn about transformers anon? Im looking into building tube based pre amps and it just seems easier, cheaper and quicker to build some to my own specs. Cheaper after an initial investment at least, ive got a couple things i could use a coil wonding machine for.

>>1104448
my knowledge is limited to power transformers. there's no good all-encompassing sources of information on the subject as far as i'm aware. more advanced matters like calculating magnetic losses and predicting coupling coefficients are extremely poorly documented to the point the answer is almost "just do FEA lol" (not to say the info isn't out there, it's just hard to find)

i don't know much about designing transformers in a context where analog signal integrity matters. give this a shot though:
http://sound.whsites.net/articles/audio-xfmrs.htm

>>1104415
Hmm if the solution is that simple, maybe I'll just open it up. Just to clarify, I'm connecting this resistor from the input to a common ground, so that the capacitor discharge current follows that path instead?

>>1104478
>I'm connecting this resistor from the input to a common ground

sorta. more generally, across the two input pins, one of which is probably, but not necessarily, grounded. a phono input, for example, will generally not be grounded, coz it needs special care.

>so that the capacitor discharge current follows that path instead

nah, so the capacitor always sees a load on the input, so that there's no significant change when you plug something in. you're getting consistency. a well-designed amp will have this built-in.

>maybe I'll just open it up

try it externally first. the problem could be something else entirely.

>>1104478
>I'm connecting this resistor from the input to a common ground

sorta. more generally, across the two input pins, one of which is probably, but not necessarily, grounded. a phono input, for example, will generally not be grounded, coz it needs special care.

>so that the capacitor discharge current follows that path instead

nah, so the capacitor always sees a load on the input, so that there's no significant change when you plug something in. you're getting consistency. a well-designed amp will have this built-in.

>maybe I'll just open it up

try it externally first. the problem could be something else entirely. e.g. it could be a capacitor-spike generated at the source, instead of at the input.

I looked up the USB connector and saw that it has nothing like a clock pin.
I don't know anything about electronics, so how does the device synchronize with the host (and vice-versa) to know when to read from the data pin?

>>1104516

it's complicated.

>>1104516
i don't know USB but UART has no clock and instead uses an agreed upon bitrate

What did bigclive mean by this?

>>1104623
Hi guys!

>>1104625
What's up you big gay bear.

>>1104626
Just enjoying some downtime. I didn't expect to be mentioned here. I typically stick to /pol/ but I had this thread open in a spare tab.

>>1104628
Ah that's cool man, roll on inauguration day! Anyhow I love your videos and website so keep up the good work.

> I didn't expect to be mentioned here

this lardass is here every week shilling his channel. and we hate him for it.

>>1104632
>and we hate him for it.
Well I certainly don't. I enjoy his content along with that Julian Ilett guy, those two should do a meet up video it would be fabulous.

I'm doing a project that requires a linear actuator. It requires 12v DC. There's a billion power supplies that provide that at sufficient amperage, but they almost all use those hollow cylindrical connectors while my linear actuator just has 2 wires coming out of it. I'd imagine it's pretty straight forward wire it up to a female connector, but I have no idea what the name of those type of connectors are. Any ideas?

>>1104661
They all look something like this

>>1104663
>>1104661
Literally "DC connector"

>>1104661
>>1104663
While >>1104665 is right, you might have more luck finding something using https://en.wikipedia.org/wiki/Coaxial_power_connector

>>1104661

Try DC barrel jack.

>>1104625

LEDs and /pol/? Are you me?

>>1104661
if you just need an easy solution, try searching ebay "2.1mm power jack breakout adapter"

What do you lads think about these?

https://www.aliexpress.com/item/AC-DC-Ammeter-Voltmeter-Ohm-Electrical-Tester-Meter-Professional-Digital-Multimeter-DT830B/32627351886.html

I need some cheap meters to measure current, im also thinking about getting this for something realable and sturdy.

https://www.aliexpress.com/item/Original-Fluke-17B-F17B-Digital-Multimeter-with-temperature-probe-can-measure-temperature/32548406195.html

>USB is 4 pins
>5 Leads come out of the board
What gives? Is one a ground?

>>1104694
OTG

>>1104674
/pol/ has the weirdest concentration of electrical engineers and electricians...

>>1104710
I took it apart, It's ground. Anyway, ground, data+, data- are delete.

>>1104722
They get shit done.

>>1104727

Specifically it's the shielding for the cable. USB requires something like a 50% wire wrap shielding for proper protocol. The shielding must be grounded, it basically works as a faraday cage for the cable and reduces EMI, both in and out which allows USB to reach its higher data rates. If you were to remove the shielding, you would significantly reduce the data rates of the cable.

>>1104773
It's a charging cable for a 360 controller. The controller itself if wireless, and the cable doesn't add wired capabilities, it just provides power.

I just removed anything that wasn't power. Not sure why it needs data lines anyway.

Anyone know good brands for terminal connectors, wires, and relays?

trying to work on my car but I want some quality stuff.

>>1104778
Dont know about wires and stuff but good relays for cars are made by Hella and Bosch. Or get the code from the relay and buy the same one as the original.

>>1104778
I recommend monster cables.

>>1104775

Oh if it's just for charging a 360 controller, yeah you're probably okay with not having the data lines/shielding.

Fun fact: If for instance this were for a phone charger, phones actually do make use of the data lines. Depending on the resistance between the D+ and D-, it denotes whether the charger is rated for "fast charging" (i.e. greater than 1A). Apple uses some specific value and Samsung shorts the two I believe.

>>1104722

I'm actually a mechanical engineer who likes electrical/electronics.

>>1105064
what industry?

>>1105073

Energy sector.

is it ok to use butt end splices on ac wiring if i make sure there is no exposed conductor on the splices, wrap the exposed section in electrical tape, then heatshrink over the tape to keep it in place? it would only be passing about 2 amps at 120v and its going to be totally stationary.

>>1104722
I can vouch for that.

>>1105212
It was quite unexpected. I posted a scope cap once and 6 or 7 unique IDs popped out of the woodwork with relevant EE bantz.

>>1105064
>1488

>>1104238
Ti, they have free samples for everything

>>1104625
ur a big guy

Suo /ohm/

Ive been going through Electronic Formulas, Symbols, and Circuits by Forrest Mims just using it to refresh on some stuff and what not.

I came across the page in pic related and was curious as to how you would determine what values of capacitors to use when doing those types of filtering.

>>1105064
This looks like the thing from Steins Gate

got a name for it or any tutorials on how to make?

>>1105471
Find the frequency of the pulsating DC and build a lowpass filter with a cutoff lower than that frequency. Say for instance when you rectify 60 hertz you get 120 hertz periodicity.
A 0.1 uF capacitor will have a shunt impedance of 13k at 120 hertz but at 1 Hz it's 160k. At DC it's well near infinity (no path to ground).

>>1105498
That's what I thought you'd do. I remember some about filter design from circuits classes, but having no resistor threw me off. Thanks.

>>1105496
they're nixie tubes. they need a 100v supply at a few mA each, which you can get reasonably cheap off ebay if you're too lame to make your own. you can control them with a few high voltage npns interfaced with a regular microcontroller.

dave jones has a recent video series where he used a 50v npn array and zeners which is worth watching if you want a project.

Rollan

>>1105546
>plasma speaker
Sweeet i just got a new flyback too

Is it a bad idea to solder the data lines together when using usb? I only need the 5V and ground wires, will the attached devices get confused if data- and data+ are connected?

>>1105617
leave the lines open, keep your shorts in your pants.

>>1105471

use 1000uF per ampere of expected load current. dont let any calculator fags tell you otherwise.

>>1105471

for power supplies, use 1000uF per ampere of expected load current. dont let any calculator fags tell you otherwise.

>>1105626
And how did you determine this?

>>1105627

observation. trial and error. experience. looking at other people's designs.
you know doing shit, and taking measurements instead of asking to be spoon-fed. one single measurement is often better than 1000 questions on the interwebs.

>>1105549
Go for one of the half bridge designs and use an LC resonant tank, monster arcs with loud music will follow.

Just remember to put a safety gap or you'll destroy the LOPT from flash over like I've done on so many occasions! The high AC voltage with destroy the internet output diode too so watch out for that if you need that diode for other drivers.

Also have you tried the ZVS driver? I got 1ft arcs from 48v input.

>>1105690
Have any links to get me started on that? Sounds interesting and fun af

>>1105634
Makes some sense. Although I feel as though it should be driven by the load's requirement for voltage ripple. If it can tolerate some just use a 0.1 uF, if it's an RF amp or something definitely needs to be as little ripple as possible

>>1105691
Dump incoming.

This one is what I used, you can get some decent sized arcs from it when tuned correctly but for "good" sound the flyback section needs around 24v or more at a couple amps.

The voltage that the transformer "see's gets halved so you only need a few primary turns around the core, I think I used about 5-6 with 36v and the difference in arc length and sound loudness was greatly changed by the number of turns.

The LC tank part is basically putting a good quality film capacitor in parallel or series with primary coil to make it "ring" up to a higher voltage, but make sure your MOSFET's can handle it. Unlike those 555 timer drivers I've never managed to kill a MOSFET with this driver using IRFP260's, use good quality low ESR electrolytic capacitors too as all of the primary current is AC couple through them and this could hurt efficiency if using a lower supply voltage.

The internet and youtube are full of people who have built this so its hard to get stuck.

>>1105719
Use the "better output stage", it does work with lower, safer +Vd voltages too at the expense of more MOSFET heating. Just adjust the number of primary turns accordingly and use about 10,000uF capacitors for upto 50v input.

The "simple output stage" does work but the MOSFET will get very hot very fast because of avalanche heating, you can try using snubbers across it but it can get complicated.

Is there any reason why a SMD ceramic capacitor would show more capacitance in one direction than the other? I'm trying to figure out the correct capacitance for some capacitors in a watch.

>>1105546
Rooling

Also I dont browse /ohm/ too much, but is arduino talk allowed here or should I wonder over to the soon to be dead arduino thread?

is anyone aware of a comprehensive list of voltage drops of various diodes? I know there are general numbers for different types of diodes, but i'm looking for something on a per model basis.

>>1105546
Rolling

>>1105798
Fuck that reroll

>>1105783
digikey lets you sort diodes by forward voltage. the actual value varies with current and temperature though.

whats a good source of analog ac and dc volt and amp meters? dont need super duper precision, just to be good while being cheap.

>>1105820
analog panel meters is the word im looking for here.

>>1105627
He didn't, it's an old rule of thumb.
>>1105634
You know, all that narcissism isn't really making you better or more convincing.

i want to make a high amp (12v 100amp) dc to dc voltage booster. ive seen a 555 based booster but its only capable of like 100ma. is there any way to make it capable of taking very high amp loads? could you separate the high current output from the driving circuit? or is it easier just to make a regular boosting circuit?

>>1105865
*regular inductor based boosting circuit

>>1105865
>>1105866
capacitive doublers are only practical to a few hundred mA. you'll need an inductive boost converter with a discrete power element that the 555 drives. you'd be better off using a dedicated boost controller though.

>>1105900
submitted early. 1.2kw is a big converter so realistically you'd just have 2-4 of pic related in parallel (and hopefully interleaved).

switching converters are difficult at anything other than low power. if you're not experienced then just get a few boost converters off ebay. they'll have terrible regulation and transient response i'm sure but that probably won't matter for whatever you're doing.

ex: http://www.ebay.com/itm/281797179064

I found a busted projector tv and was looking into the parts.
The dlp chip looks good which is the most important part. How hard would it be to turn this into a projector. I know how to program though as far as firmware Ive stuck to arduinos. I have an idea of how signal processing works but never looked into it much.
Basically the way a projector works is I tell the chip what mirrors to turn on, the I shine a color of light on the chip and send the reflection through a lens. The color wheel seems easy enough to control, but Im not sure what would be required for the chip.

what is a good way to isolate circuits using the same power source?

I have a 1500w PSU

I also have 2 main loads
>Load A
sensor and microcontroller network (<300W)
>Load B
led array (700W)

I am dropping the voltage down for load A and do not want any fluctuations in load B's power draw to interact with load A.

what is a good way to isolate circuits using the same power source?

I have a 1500w PSU

I also have 2 main loads

>Load A
sensor and microcontroller network (<300W)

>Load B
led array (700W)

I am dropping the voltage down for load A and do not want any fluctuations in load B's power draw to interact with load A.

>>1105936
for some reason the 4chan text area is reading my tags and commenting out text.

load a is less than 300w

>>1105936

the only surefire way is to have separate supplies. but if you insist on using one, then you need to
- put the more sensitive load closer to the supply - i.e. use longer wires for the LEDs, or make a coil of the wires, or add a heavy duty coil in series.
- add like 50,000uF caps near the LEDs to absorb transient spikes.
- put a heady-duty diode before the cap if you can afford the voltage drop

>>1105725
To elaborate, this is a 30 year old capacitor in a Seiko G757.

>>1101963
Add NEETS to your list:

http://www.fcctests.com/neets/neets.htm

What's the probability of me killing myself while messing around with 12v dc? I have no idea what I'm doing, btw.

>>1106066
>12v dc

Just unplug it before you work on it, m8. Keep a fire extinguisher nearby in case it ignites when you plug it back in.

>>1106066
If you're not standing in the rain, practically zero.
You can stick your tongue to a 9V battery safely to test it
If you're dealing with a supply that can produce a lot of current (like a car battery), the biggest dangers are burns, fires, and bits of molten metal in your eyes.

>>1105506

You can get everything off of ebay including the high voltage binary decimal converters. I bought a power supply for them because I don't have an oscilloscope yet.

Can any recommend me a core for wrapping my own inductors?

Looking for a core both suitable for low frequency applications like audio filters and for high frequency oscillators.

Will also take separate recommendations on core types for power circuits like boost converters.

Is 22 AWG enameled copper wire good for low power inductor/transformer winding or no? Is it too big?

>>1105546
rollingo

I'm trying to figure out a way to control an universal motor with triac's. Is this going to work at all?

I recently got myself a more sophisticated function generator and had two questions. What's the difference between the balanced and unbalanced outputs? And why is the output amplitude double the displayed amplitude (this has something to do with the output needing to be terminated into the correct load corresponding to the set output impedance but I don't fully understand this)?

>>1105931
Won't be doing anything with it without finding a datasheet for it first.
And then getting the chip seated, wired and interfaced is going to be hell.

>>1106200
Balanced: puts out signal and signal*(-1) on two wires. Resistant to common mode noise/interference.

Unbalanced/single-ended: puts out signal on one wire, referenced to ground/0 on other wire; generally ground is shield.

>something to do with the output needing to be terminated into the correct load corresponding to the set output impedance
Right. For 50ohm out it expects a 50ohm load. There's 50ohm inside the box and you provide 50ohm outside, so say for 1V output the box will make 2V inside. If you hook up 10 Mohm outside you will see the full 2V.

>>1106200
what is the deal with that multimeter brah

>>1106221
If I put a 50 ohm load in parallel with a 10Mohm load will it work properly?

I can get one of those BNC T adapters and a 50 ohm terminator and hopefully solve that problem.

>>1106250
Yep. Or just run it at 1/2 amplitude.

Considering this an ideal non inverting amp, could anybody explain me how to solve for Vo?

Whats confusing me is the second voltage source placed on the non inverting part.

I am not quite sure on how to solve the problem now that two voltages on two diff terminals are involved.

Although I know that assuming linearity means In = Ip = 0 and that Vp = Vn. Still its not clear.

>>1106344
2k -> short
10k -> open
vb is 1.5v relative to va
vo = 1.5v*(1+40/5) = 13.5v relative to va
vo = 14.5 relative to ground

proof: 13.5v*5k/(40k+5k) = 1.5v at inverting input, same as noninverting

honestly if you don't know how to solve these just put them in a simulator.

>>1104347
i find that my fluke clamp meter will read something with the jaws open and no current flowing however this is irrelevant really because we all know that these things wont work with the jaws not in the closed position.

For when its not zeroed and the jaws are closed i normally physically hold the meter away from any sources to see if it goes back to zero and if not it has a zero button to bring it back into line.
Maybe yours has a zero function too? any mentioning in the manual?

I want to build my own power bank.
I want it to be as efficient as possible.

Can I use an ic like maxim MAX1709? Are there any better, similar priced ICs out there? Can I use parts from dead computers for it?

Is it safe to order these ICs from aliexpress? Am ausfag, difficult to find shipping here.

Thanks. This is going to be my first project so I'm super grateful for any help.

What's the deal with 110VAC, 115VAC, and 120VAC in the USA? Are they all referring to the same thing and ignoring the ~10 volt variation?

>>1106379
the actual voltage is 120+-10% (or better). mine is 114-126 for example and currently reads 122.

how do i calculate how much current a group of wires are capable of? for example lets say 10 gauge has a max amperage of 50 amps, so if i braid together 3 wires it should take up to 150 amps right?

>>1104347

It's a shit meter. It'll work fine for anything you'll use it for, but any precision measurements will require a different tool. And probably a lab.

>>1106183
You can't tie TRIAC gates together that way. They'll die.
I guess you could make the basic idea work, but rectifying AC first and then operating on DC is probably easier.

>>1106416
Well that picture is just a basic concept, it doesn't have DIAC's or opto-couplers, so the gates won't be tied together. I can, for example, full open one TRIAC and then half open the other. I was asking if the basic idea would work and if you can put a load between two TRIAC's.
Because if not, yes, I'd just rectify and build a H-bridge. Besides, wouldn't that also make the motor run a bit more efficient?

I'm a stupid moron and plugged in a wrongly wired power supply into this POS screen (+/- switched). Now it only flashes the backlight when I plug in the correctly wired power supply. What do? I have means to replace any components, but I can't into electronics to find the problem. Where should I start?

>>1106401
>it should take up to 150 amps right?

You'll lose a little capacity due to bundled wires not cooling as well.
Also, wires aren't perfect and one could end up taking more current than the others, so you lose a little capacity there too.

But basically, yes. Just allow for a bit less capacity than just adding them together.

Why do we have two threads?
Why are you guys using the younger thread?

>>1106452

any board that received reversed power is probably dead for ever. there's a small chance that some fuses blew and protected the board, in which case you can bypass them. they tend to look like larger versions of black SMD resistors, often with F marked on them, and PCB marking showing a part number like F1 1.5A, or F2 3A, etc. test for zero ohms across them with a multimeter in the ohms range.

I usually only used premade relay boards that are connected via jumper wires to whatever I'm building. I now plan to put relays on a PCB myself.

Anything I need to know or can I grab just any relay and slap it on a board? I'm mostly concerned about blowing up the PCB because of wrong trace width or anything else I might not know yet.

There's no specific use in mind, it's just going to be a general purpose board with a few transistors and relays connected to an ESP that I later may use to remotely trigger the press of a button or light up an LED.

>>1106491
wtf i hate dc now

>>1106491
>not connecting a polarity reversing switch to your mobo so you can run your PC in reverse

ISHYGDDT

I have one of these continuous rotation potentiometers being read by a microcontroller. It produces noise when the wiper is somewhere around the 0/360 degrees position, presumably because the wiper is making contact with both the high and low resistant ends of the track. Is there any way to fix that or is a dual wipe pot the only option?

>>1101963
so I was working on a little project just now and I dropped my soldering iron at 400c on my hand for just a second.
How long does it take to get a 3rd degree burn? I didn't feel any pain and I've got a white patch of burnt skin now
It's pretty small

>>1106738

I've burnt my hand on an active oven coil and got the same thing. It's probably a low 2nd degree burn. It will start hurting soon.

>>1106738
Rub your hand on your hair. Trust me. No scar left and almost no pain.

>>1106738
You'll be fine, just clean it regularly and keep it wrapped

>>1106738
Get a knife and cut out the burned flesh. Stitch the hole up. Done.

>>1106738
i burn myself all the time. it's never left a scar.

i have brushed my iron in brass wool and flicked a little solder bead right onto my eyelid though so i'm a little careful when doing that.

>>1106738
your skin will have a pinchable waterbubble soon :^)

>>1106401
Pretty much sum the ampacity, yeah. Just make sure all your conductors are EXACTLY the same length and terminate them properly (no wing nuts)

>>1104070
>The screw is behind the front plastic and there's no way to open it.
It's not a screw, it's just a rotatable insert in the front plastic.
On the back (inside) of the insert is a small off-center pin that fits into a slot on the meter coil spring adjustment.
As the insert rotates, the pin moves the adjuster right or left changing the position of the meter pointer.
Use a needle to pick out the red sealer covering the slot in the insert and around the edges so it can be moved.
Don't use any solvent. It will weld the plastic.

>>1105203
yes

also staggering the splices is recommended.

------------------------===-------------
--------------===-----------------------

>>1105546
Rolling

>>1106828
Rerolling because who the fuck would want to build a music box

>>1106798
>>1106796
>>1106765
>>1106753
>>1106748
>>1106743
Nothing has happened yet
doesn't hurt
still a white spot
there is a little red around the outside of ground zero but that's it
I can poke it with a fork and not feel anything but a little discomfort from around the edge where it is red
I do have it wrapped now though

Hey guys, i received a credit card reader from a friend shop closed down and had no need for it.

device
>verifone vx510

I never took one apart, and I knew i would trip anti tamper features on it.

I mainly wanted
>pic related

I was able to find somehwat of pin out.
http://2.bp.blogspot.com/-1Px98z-du3c/UBfaneFkoWI/AAAAAAAAAOk/ondFKhNsRqg/s1600/head.png

can I hook reader head directly pins headers on arduino leonardo and be able to use it, or will it be impossible or too complicated and be better off to get usb card reader? I essentially want to use card swipe to lock/unlock doors.

>>1107115
Sorry forgot to mention pic related is magentic card reader head (3 track)

hey /ohm/

I'm trying to make a USB HID game controller.

I have some basic electronics components (buttons I'd like to use, breadboards, wiring, good soldering iron, resistors, etc)

I also have a
TI MSP430 LaunchPad
TI Stellaris LaunchPad

Having a bit of trouble trying to figure out where to start.

Which micro should I be using? Any general advice for working with TI launchpads or for this project in general would be appreciated.

// posted again because the other ohm thread seems ded //

Has anyone in /ohm/ put together a buy list? I wanted to put together a work station starting from scratch, and I'd like a tiered list of different versions of equipment to compare them.

>>1107118
>MSP430
I was just looking over my projects from about 4 years ago.
Forgot most of the stuff, but thinking about getting back into it.

Here's Info on the MSP430 Clock. Block diagram, control registers, bit explanations, and relevent header aliases all in one pic I made. (Was also learning inkscape at the same time).
Hope someone else finds it useful.

Does a polarity reversing switch literally just reverse the flow of electricity?

>>1107128
Neat. Did you use CCS when doing 430 dev? I didn't know it costs like $500 to program non-dev chips with it.

So I'm looking at using Energia now... I don't know if Energia supports making the Stellaris into a USB HID peripheral

>>1104308
Question.. what is magnet wire insulated with, I've always wondered.

>>1107142
Yea, CCS
Here's a screencap from when I asked for help.
Program was for IR codes.
(Problem was solved by dividing first because it was overflowing int)

>>1107143
its a flexible, painted on lacquer basically.

>>1107132
What else would it do?

>>1107150
So like this stuff, just a higher quality and applied in factory.

http://uk.rs-online.com/web/p/electronics-varnishes-lacquers/1181554/

>>1107156
idk about higher quality. the methods and materials used to lacquer magnet wire on an industrial scale hasn't really changed since WW2. i wouldn't be surprised if you could do up your own at home with some spray can type stuff. i know that they have lacquer patch kits for motor windings.

Are there any affordable oscilloscopes?

>>1107163
what's affordable? if you have a proper job just get the ds1054z for $400 (or the 2ch equiv for 350 iirc). if not you can get $50 analog scopes.

>>1107163
Depends on what you need it to do. There are always old scopes on eBay.

>>1107166
It's not a matter of having a proper job, $400 dollars is a little pricey for a single piece of equipment for a hobby...

>>1107168
your options are analog or usb scopes then. everyone says usb scopes suck but i've never used one and i don't think they have either, so don't neglect that option.

i've heard that the eevblog forum arranges free analog scope giveaways to new hobbyists so that might be worth investigating if you can't get one off ebay.

>>1107163
https://www.amazon.com/JYE-DSO-138-Open-Source/dp/B00WAQGGZA/ref=sr_1_4?ie=UTF8&qid=1483230808&sr=8-4&keywords=dso138

>>1107143
>I've always wondered.

I've always wondered why people don't use google first.

https://en.wikipedia.org/wiki/Magnet_wire#Insulation

>>1101963
can someone tell me why theres capacitors all over every pcb I see. Like what purpose do they serve from a computer science perspective. Are they to regulate timings?

>>1107204
digital ics consume current in bursts. pcb traces have resistance and inductance, so these bursts cause voltage drops that will result in the ics malfunctioning. adding capacitors near the ics counteracts this by providing localized energy that can be used in the short term and refilled in the long term by the main supply.

that's usually why they're there, but obviously not the only reason.

I have a dev board that has some GPIOs. What I want to do is hook up push buttons so that when I press a button, it lights up the board LED different colors (can control w RGB)

But I think there are more buttons than IO pins. What do? I have no idea what I'm doing btw, first foray into electronics

>>1107208
multiplexing/charlieplexing. examples from google images.

what method you use depends on whether you have the spare processing power to scan inputs vs using interrupts. you can set the scan up as a relatively slow interrupt routine so it's not as big a deal as it sounds.

Are ebay copper clad boards bad?
I bought some ultra fucking cheap single sided boards that I plan on making small things with.

Where can I find info on analog video signals?

>>1107216
I'm kind of an expert on this subject. I do a lot of analog video synth work. I can answer some questions if you like.

https://www.youtube.com/watch?v=CRSLbo_8nTQ
Anyone have any experience with automated pcb placement.
I have a hot air gun, 5 nema17s, some arduinos, a pi zero, and a 3d printer.
Other than some threaded rods, syringes, and a few cheap stepper motors, what else would I need?

>>1107219
Hard to narrow it down to specific questions. I'm interested in converting digital HDMI video signals to analog composite and YPbPr component for display on old NTSC CRT televisions. There's lot of information on the digital video format so that shouldn't be too challenging to figure out, but I can't find much info on what an analog video signal actually looks like. Know any good documentation on the topic?

Can anyone refer me to a pinout/internal for a 3pdt switch? I want to make an audio output switch but EVERY SINGLE search result is guitar stomp boxes. I'd buy a switch if they werent thirty god damn dollars. Thinking of tying in the switch for my powered speakers at the same time. It safe to put 120VAC through a soldered rocker switch?

is it standard for microcontrollers to have internal pull ups / pull downs?

is it safe to connect a push button to a gpio without a resistor and a debounce cap?

>>1107270
pullups yes, i don't think most have pulldowns though i definitely know at least one pic does

yes but software debounce is more annoying than an RC

>>1107257
>pinout/internal for a 3pdt switch?

Is this bait?

>>1107257
>It safe to put 120VAC through a soldered rocker switch?

It depends on what the switch is rated for and what the solderer's skill level is.

>>1107279
I apparently missed that image. I wanted to make sure because some switches can be rather confusing. For the 120v, as long as the switch itself is rated for the power being put through it and the solder contact is thick enough to support the current, its all good? I might just do a 4pdt and bundle headphone/speaker toggle and speaker power all on one switch.

>>1107293
heatshrink the connections to prevent your drool from bridging them

>>1107294
hurr

>>1107295

>>1107293
>the solder contact is thick enough to support the current,
You should get someone with knowledge of solderfication techniques to do this for you.

>>1107293
>solder contact is thick enough
solder thickness is not a measure of conductivity or connection strength

in fact overly thick (and non-shiny) solder is probably an indication the connection is not very good

the solder should be tight and shiny and flow across the connection points easily when you apply it

>>1107316
Is it not the same cross-sectional area concept as wire gauge?

>>1107320
The conduction should take place primarily between the conductors.
Not through a blob of solder.

Find a course on 'How To Solder' and follow the instructions.

>>1107323
https://www.youtube.com/watch?v=zusJSKeXGHE

Ok I am looking for ttl magnetic card reader, and to be honest
>pic related

has been the only one to provide results.
Anyone else shop for parts, and have you been scammed, or no?

Can anyone point me to a cheap crimper that can crimp uninsulated terminals for 10awg? I'm trying to crimp Molex Super Sabres terminals.

Happy new year h/ohm/os.

>>1107408
am i an asshole for hating this comic? nothing to do with munroe.

happy new year :3c

>>1105546
Author of this (and author of the original op) here. Thinking about making an updated version with easy medium and hard choices for each roll.

Anyone have any suggestions on what I should add? Otherwise I gotta think of 200 more projects on my own.

>>1107413
do a nigglet/pirate edition, card skimmer/atm/id/hotel whatever.
rfid reader/write 125khz

stuff like that idk its almost 4 am here.

>>1107416
Maybe, as long as we all pretend it's for educational purposes only

>>1107417
>>1107417

do you guys think I will be okay if I use a toaster as a reflow oven without any auto temp. controlling stuff? (ex. I'd be monitoring the temperature and adjusting to create profiles manually)

>>1107383
>crimper that can crimp uninsulated terminals for 10awg

Gardner Bender GS-388

can be found online or probably at a local hardware store in the electrical department

~$10

>>1107491
Looking at the reviews it looks like a horror show. It looks worse than the chinese crimpers I have, that unfortunately aren't big enough for 10 AWG.

>>1107501
Oh I forgot. The Super Sabres are open barrel crimps

>>1107408

Happy current year!

the other day i finally manned up and fixed a marshall tube amp. however i got an electric shock because i was soldering it while plugged and running lol

>>1107411
Everyone hates it these days. Protip: if you want to stay successful, never mix political agendas with work.

>>1107229
This is pretty good documentation on an NTSC signal.
http://belogic.com/uzebox/video_primer.htm

It's also very useful to understand that NTSC sync and porch timing matches a 640x480 VGA signal except the number of scan lines between vertical porch and sync are divided by two due to interlacing.

>>1107501
>It looks worse than the chinese crimpers I have
You asked for >>1107383
>cheap crimper

>>1107501
>aren't big enough for 10 AWG
They do 8 AWG uninsulated

>>1107208
>>1107211

Missing the point:

you don't need a computer for that

Just wire the buttons between V+ directly to the LED terminals (unless you are pwming them)

>>1107587
>>It looks worse than the chinese crimpers I have
>You asked for >>1107383
>>cheap crimper
I meant $20-40, not scrap

>>aren't big enough for 10 AWG
>They do 8 AWG uninsulated
I'm talking about the ones I have

I'll just get one of those Greenlee crimpers next month.

>>1107433
I've done it. My oven tended to run hotter than what the thermocouple was reading, and I never bothered to figure out while. I recommend doing test runs.

semi-related..

Are cement-cased resistors safe to solder to a spliced wire without any additional housing?

So /ohm/, I'm trying to build a basic AM modulator, but for time codes, not for radio.

I have a source that outputs 3.3V when high, and a 3.3Vpp 1kHz AC carrier wave.

When the source is high, it needs to be 3.3Vpp 1kHz wave, and when it's low it needs to be a 1.15Vpp 1kHz wave.

I've tried toggling a drain BJT and using an inverter, and I get the voltage reduction but it cuts off the bottom of the wave. Would a voltage divider work when just a resistor doesn't?

I've tried googling around, but all I can find relies on an AC input from a mic to a transformer, but I have a DC input.

Is there an easier way to do this than going from my D.C. Output to act then into another set of transformer coils n phase with the first set?

>>1107831
Those things get very hot, if you run them anywhere near their rated power. Another thing is that they aren't designed to handle random stresses on their wires.

Thus, one of those randomly dangling on a wire isn't the greatest idea, particularly if it's allowed to move.

Hello, I want to build an O2 sensor output modifier using an Arduino.

The sensor works by giving more voltage when the fuel mixture is rich and lower when lean, the voltage is between 0.2-0.8v.

My idea is: I will read these voltage (probably by amplifying it first to be readable by the analog input) and outputting a fake one to the ECU.

My question is: how do I output 0.2-0.8v using Arduino?

>>1107870
You don't need an Arduino to do that. Just amplify and bias the sensor voltage with a single op amp.

>>1107874
Unfortunately, I think I need it for this project as it would read data from other things, not just exclusively doing only this.

>>1107874
Also you can just choose suitable rail values to cap it at 0.2-0.8V.

This is assuming linear amplification is acceptable as you haven't told us the desired response characteristics.

>>1107876
Unless you need to adjust the output based on previous data, using an opamp will give better results.

Arduino can't output steady DC values off it's analog pins. The best you can do is LPF a high duty cycle square wave.

You can still use the arduino to capture data while keeping the sensor/output circuit separate.

>>1107879
Hold on, elaborate further about your plan. Could I still control it using the Arduino?

>>1107880
It's called PWM (pulse width modulation). Just Google it, there should be tons of tutorials.

>>1107876
Use more op amps as comparators. Still no need to go arduinolol mode

>>1107882
OK, I know about PWM. What term should I search for this op-amp control?

>>1107884
Are you dead set on using an Arduino?

You haven't specified what you're using the analog output for, so a filtered and buffered square wave from the arduino may not be acceptable at all.

>>1107887
>Are you dead set on using an Arduino?

It is what I currently have right now

>You haven't specified what you're using the analog output for,

for mimicking the output voltage from the O2 sensor

>so a filtered and buffered square wave from the arduino may not be acceptable at all.

That was why I asked if somebody could spot a flaw in my original plan. And no I don't have an oscillator to see what the waveform is from the sensors. The only thing I know is it is feed directly back to the ECU.

>>1107888
>It is what I currently have right now
You're going to need to buy extra components either way.

>for mimicking the output voltage from the O2 sensor
still confused about this part

>That was why I asked if somebody could spot a flaw in my original plan. And no I don't have an oscillator to see what the waveform is from the sensors. The only thing I know is it is feed directly back to the ECU.
Depends on the equipment that's reading the 0.2-0.8v output.

>>1107890
>You're going to need to buy extra components either way.
I agree

>still confused about this part
So currently I am planning to create a turbo spooler for my car using compressed air, but it will fucked with the O2 sensor and throwing it off because of more oxygen. I want to modify it's output to the ECU to pretend I don't inject anything to the exhaust.

>>1107893
>a turbo spooler for my car using compressed air,
https://www.youtube.com/watch?v=nBPeQG40sI4

>>1107893
>but it will fucked with the O2 sensor and throwing it off because of more oxygen
Only momentarily (while the air is injected) and may not change it as much as expected.
If you are already having problems with it after installation, I can see looking for a solution.
If you are still in the planning stage, install it first to see if it's really a problem.

I want to get into smd pcb. I dont have the money for an oven, Is a hot air gun enough?

Hey /ohm/ I ordered a card reader that uses RS232 connection, and I am receiving it this Friday, I ordered some rs232 to TTL UART for my aruino leonardo, is it possible to nig rig a re wiring until my converter comes in on Monday?

>>1108760
There are plenty of SMD components you can solder with a normal soldering iron.

>>1108762
You can do RS232 to TTL conversion with couple of transistors and a handful of passives, if you don't need much speed. Pic related.
It does not handle the handshake lines, but there's a decent chance your reader does not use them. If it does, you can try shorting pin 7 to 8 (RTS to CTS) and tying the remaining 4 pins together as well.

>>1108766
I am new to reading schematics the multi line triangle thing at the bottom, what does that represent? ground?

>>1108769
If you're referring to the component marked DB9, then it's a D9 connector, commonly used with RS232. Pin 5 is ground, 2 transmitted data and 3 received data.

So I recently got into the book about the basics of electronics and all. I have a question about transistors that I can't seem to find an explanation that clicks with me, how exactly does it amplify a current, does there need to be two sources of power for that to happen?

>>1105546
roll

>>1108775
>I can't seem to find an explanation that clicks with me,

Small force on base controls large force between collector and emitter.

>>1108775
>does there need to be two sources of power for that to happen?

Sometimes two are used but generally the small control current comes from the same source of the controlled current.

>>1108775
Man I don't remember shit about this but

>how does it amplify a current
Its a property of the materials used to make the transistor. If you want a more clear understanding, you need to search for "silicon doping" "NPN diodes" "NPN transistors" "animation" etc to see what's going on when quantum theory is generalized to electrons and holes and stuff

Damn I feel bad that I don't remember *any* of this

>>1101963
i bought some 100 watt led cobs and jesus fuck theyre bright, like i have to wear my welding helmet bright. the specs are 100 watt, 32-36 volt, 3 amps max, and at 35 volts they draw about 2.8-2.9 amps (98-101 watts), so pretty much exactly what i paid for. however they get fuckin hot, after like 2 minutes you can put your fingers an inch over the cob and it will be like holding it over a hot incandescent. my laser thermometer clocked the led array itself to be about 260-280f. it is thermal pasted to an old large cpu heatsink, nothing super fancy but its large and moves a ton of air. it stays in the 260-280f range while the heatsink is on but as soon as i turn it off it gets cool in literally a second. is this too hot for continuous usage?

>>1108860
also i forgot, my house supplies 125v, and though a diode bridge i end up with 120v, can i take 3 or 4 of these cobs and put them series to divide the voltage to each?

>>1108775
It's not really an amplifier, it's a voltage-controlled variable resistance. The amplification comes from what you put around it.

I love how 9/10 people(I'm the 1/10) in this thread have no idea of what they're actually doing. They just think "hurr connect resistor to get more power durrr"
They have no idea of the fundamental physics that lies behind these electronic devices. I doubt any of you fags have ever even simply seen a textbook on electricty and magnetism let alone even taken a physics class.
None of you should even look at a circuit again until you actually learn what you're doing.

>>1108827
>>1108832
>>1108877
I'm pretty sure I understand it now, I think I was looking too far into the "amplify" part with a literal meaning . Thanks anons!

>>1104625
Big Clive, What a legend

Can anyone recommend a place to get a half decent solder sucker? I bought a cheapo one and the top of the fucking thing flies off half the time when I try to use it

>>1109016

search google for a store that sells electronics components in your town. if you find one with a website that has a picture like this, that's the place to go. otherwise, order a name-brand unit online from digikey, jameco, etc (all the stores listed in the sticky at the top)

can we make a thread about HOW FUCKING AMAZING MICROCONTROLLERS ARE. I recently got introduced to the pic family. and the design to build time is awesome. especially compared to arduino. Pic is just simple and beautiful. But arduino is still great for complex projects and pretty insane when you think abot alll the power it has.
(pic semi related)

>>1109475

no we cant. we have already bullshit threads about why Bosch is better than Makita, etc. one perpetual circle-jerk on /diy/ is enough.

>>1109483
>Bosch
>Better than Makita
Now if you said DeWalt, maybe.

>>1107939
yea, that is what I am trying to do, diy style. I haven't installed it yet, but I guess you are right, I have to try it first, maybe it will just fucked up and bent my valve :^)

>>1108860
100W for CPU heatsink that is rated at (usually) 65W TDP? And you ran it without fan? No wonder it got hot m8.

>>1109475
>and the design to build time is awesome. especially compared to arduino
What do you mean? Compiling takes ages and there's no great difference in the programming.

Interpreted MCUs when?

>>1109492
>>1109483

You're both wrong.

>>1109580
Yep. Makita is great except for the god awful color scheme

>>1109573
Takes ages? what is this the 80's?
>Interpreted languages
On something with 256 kb flash and RAM?
How about never.

>>1109573
>>1109616
PICaxe not PIC. My bad.

>>1104931
Fun fact, so does that charger. It will only work when connected to the console

trying to fix my flourescent lamp wtf am i doing wrong? this is how i wired all up only top tube lights up and if i swap the tubes around same thing happens, i tried replacing the bottom starter but didnt help

>>1109856
You left out part of it.

>>1109906
the ballast is there my circuit starts after the ballast

>>1109911
need a twin ballast or separate ballast for each tube
when the first tube starts, the voltage drops across that tube
since you have the tubes in parallel, the second tube never gets enough voltage to start

>>1109918
Basically i got this old lamp which has a ballast a weird switch with what i presume has built in starter but its using metal plates and it managed to start both the tube. But the switch broke and its beyond repair it had springs bronze/copper plates and what not and everything was connected thourgh it. So since i cant repair it i started trying to substitute it with modern starters and the same old ballast and i cant get further than to light one tube. How did the old ballast light both without 2 starters even?

>>1109918
>>1110050
picture of the ballast + my experimentation

>>1110050
>How did the old ballast light both without 2 starters even?
The damaged switch was the 'starters'
You replaced it with 'automatic switches' which operate by the gas inside the starter glowing and closing a bi-metal (heat sensitive) switch.
The closed switch puts current through the filaments for a moment making the mercury gas inside the fluorescent tube easier to ionize and begin operation.
Once the fluorescent tube is working the starter gas doesn't glow (the voltage across the tube has fallen below the gas ignition voltage) so it quits closing the thermal switch inside and the filaments no longer carry the heating current.
Since your fluorescent tubes are parallel, the voltage across both tubes fall to the 'running' voltage once either of them starts.
The starters are part of the same parallel circuit so when the first tube lights up, neither starter tries to close it's switch again.
Replace one of the starters with a switch (preferably a momentary push button)
Allow the first tube to operate automatically.
Push the button for a moment and you should see the filament in that tube glow.
Release the button and see if the tube starts.
I personally have never seen a two tube lamp with a two wire ballast.
All I've ever worked on had either a three wire ballast or two two-wire ballasts.
I'm not saying it can't/won't work - just that I've never encountered it.

>>1110075
changing one of the starters to a switch didn't start both tubes

>>1110050
>How did the old ballast light both without 2 starters even?
The damaged switch was the 'starters'
You replaced it with 'automatic switches' which operate by the gas inside the starter glowing and closing a bi-metal (heat sensitive) switch.
The closed switch puts current through the filaments for a moment making the mercury gas inside the fluorescent tube easier to ionize and begin operation.
When the switch opens (the closed switch killed the glow/heat) an inductive kick-back from the ballast along with the heated cathodes allows the tube to start.
Once the fluorescent tube is working the starter gas doesn't glow (the voltage across the tube has fallen below the gas ignition voltage) so it quits closing the thermal switch inside and the filaments no longer carry the heating current.
Since your fluorescent tubes are parallel, the voltage across both tubes fall to the 'running' voltage once either of them starts.
The starters are part of the same parallel circuit so when the first tube lights up, neither starter tries to close it's switch again.
Replace one of the starters with a switch (preferably a momentary push button)
Allow the first tube to operate automatically.
Push the button for a moment and you should see the filament in that tube glow.
Release the button and see if the tube starts.
I personally have never seen a two tube lamp with a two wire ballast.
All I've ever worked on had either a three to five wire ballast or two two-wire ballasts.

>>1110082
Sorry, I realized I'd left something out and hoped to change it before a response.

As I said, I've not seen a two-wire dual tube lamp before.

>>1110084
>two-wire dual tube lamp before.
two-wire ballast

>>1105546
roll

Hey, i know sex toy related things shouldnt be on here, but whats the simplest and easiest way to make the electronic components for a fucking machine? All i need is a low RPM high torue motor the has speed control.Thinking about getting a cheap hammer drill and hooking up a pot where the trigger goes.

I own ''Electronic Fundamentals'' 2nd edition by Floyd, it covers circuit from capacitors to transistors and RC and RL signal and such stuff, it seem to cover quite a lot with a focus on practical use (such as repairing). The question is: is this knowledge sufficient to build your own circuits and devices like transmitters or do you need to be an engineer for that?

>>1110163
rip apart a cheap or free treadmill from craigslist you degenerate

>>1110167
Shit thats too much work. you aint getting that shit in my market.

>>1110164
ARRL RF Exp Handbook. But you'll also need the right gear to measure your shit.

I have a chink tablet that will no longer charge.

I'd like to disconnect the battery and power itmoff the wall. Is this possible or too much effort?

It's a x5-z8500 with 4gb ram and 32gb emmc so I could definitely put it to use.

Halp?

>>1110245
Im assuming you have a usb wall plug, and a few usb cables lying around. Doesnt need to be any particular type, old printer cable will work. Im also assuming the battery is around 3.7 to 4.2 volts. This should work and wont involve taking the tablet apart.

Chop the non usb end of the cable off, strip back the shield, you can also chop the data wires cause you wont need them. Strip a little off the 5v and ground wire. Touch them to the proper terminals. There should either be a marking on the tablet or the battery indicating which is positive and negative. Connect 5v to positive and ground to negative. If it powers on you can figure it out from there, if not more than likely the battery needs to be in place for it to turn on. Do not hook your battery up to the usb cable, it could overcharge and potentially explode or cause a fire. Although, if you were to get a charge controller for your type of battery you could build an external charger for it. You may be able to find another cheap battery thats identical and have 2, so you can charge one battery while using the tablet. Otherwise an external charger will be very inconvinient instead of only mildly.

Or you could just fix the charging port, but if youre asking here about it then youll likely have difficulty with that. My idea is a quick ghetto method where youll only at worst destroy a usb cable.

>>1110255
Also, id use a diode to drop the voltage a little on the usb cable, to make ot a little lower than 5v. Radio shack has them. 1n400x(probably a 3 or 7)

Where's the best place for simple diy audio interface tutorials?

>>1105546
Rollandow

Ok guys I received my card reader, and my rs232 to ttl uart converter, I didn't pay attention and realized my convert is also female plug (needed male plug)

My initial goal was to get rid of the rs232 cable and solder on the board to straight to ttl uart for my arduino board.

>pic related

I squared some area where I might be able to achieve that,
the square that has thicker frame is what I need, but the
>int0
>xtol1

I am not sure what those are.

>>1110388
is that I think I need****

>>1109815
Thats not true, i use mine plugged into my pc all the time and it works

>>1104722
I see what you did there

Sup /ohm/.

>fan heater
>smells funny
>30 seconds later trips breaker
>plug is melted into socket

Is this fucker safe to rewire? If so I'll just stick another plug on and carry on using it.
Or should I just fuck it and bin it? It's some cheap shit that came with the apartment I live in.

>>1110498

sounds like you have multiple problems going on. the breaker going out could mean a tired breaker, or a short inside the heater. the smell suggests the latter.

the melting could mean arcing in the socket caused by oxidized contacts in the socket, or the plug, or both. arcing would mean the heater is not getting enough constant voltage to get hot, which would have increased the resistance of the element, and thus bring down the current after a few seconds.

recommendation: like Sherlock, use process of elimination - replace plug and try a different socket.

>>1106344
i passed that class but i don't remember any of it

Is playing around raspberry pi projects a good way to learn electronics?

>>1110592

it's like asking if listening to the radio makes you a better musician. if you want a deep understanding then you gotta start with the theory, like The Art of Electronics by Paul Horowitz (book in the sticky).

if you just wanna build stuff by parroting other's work, then the Raspi and Arduino are good for that.

>>1110595

You're kind of full of shit. The Arduino got me into electronics and helped me understand enough essentials to design my own board.

>>1110592
Most raspi projects are more computery. For electronics, you want Arduino.

Going to pick up a few LED strips for some projects. Is there anything I should look for or avoid on them? I've never really shopped for them, I'm not sure how different they could really be. Need a couple of white ones for a couple small projects, going to get either a white or RGB one for some extra room lighting.

I'm new to electronics as well. Glad to have found this thread. Just a few further resources I found helpful so far:
Simulate circuits in your browser:
http://lushprojects.com/circuitjs/circuitjs.html
YT:
https://www.youtube.com/user/humanHardDrive/videos
https://www.youtube.com/user/sutty6/videos

Also I'm going to buy most of my starting stuff from Aliexpress and was surprised to not find it in the list of suppliers.

Whats the difference between an LM7805 and a L7805CV?

Im not ohm savvy, It calls for an LM7805 but I already own a bag of L7805CV and am too stupid to figure out if they are interchangeable or not.

>>1110682

me again. I actually have a question regarding oscilloscopes. I will be working mainly on audio components and I would like to have a basic scope (so I don't need a logic analyzer). Now I'm a cheap ass and don't want to shell out 350€ or more for an oscilloscope at this early stage of getting into the hobby.

I've read that you could use a USB soundcard as a basic first oscilloscope. With 192kHz sample rate you are not going to get too much out of it, but for some audio applications it's enough.

Now I thought maybe there are sound cards with higher sampling frequency, but there aren't. Instead I found:
http://www.netams.com/~anton/usb-adc/
which uses a 12MHz chip as far as I can see, so that would be quite a significant improvement and with
http://catalog.datel.com/item/data-acquisition-product-line/sampling-a-d-converters/adsd-1420s
you could go up to 20MHz at 14bit resolution.

There's a wide variety of software tools to make use of the input and as I know my way around programming that should not be a big issue.

So what do you guys think of this usb adc oscilloscope idea? Will it work or am I missing something? As said, I'm a noob, so this is very likely.

>>1110510
Thanks. I'll have a go at figuring out what the problem is.

>>1110686
I would guess that they have different maximum voltage/current ratings.

One will be able to take higher input voltages & input currents.


Take a look at the data sheets to see what is different.

keep in mind that putting in a higher voltage into a linear regulator means more energy needs to be dissipated as heat.

Vout + 0.7V would be ideal as an input voltage for efficiency.

>>1110687
>USB soundcard as a basic first oscilloscope.
If you're desperate and your requirements are very low, yes, that's an option.
>which uses a 12MHz chip as far as I can see
The maximum sample rate is still 48kHz.
>adsd-1420s
Dude, that requires lots of work to turn it to a scope. It is certainly not novice's project. And, if you decide to go that route, there are plenty of other ICs from TI, Analog, etc.

>usb adc oscilloscope idea?
They kinda suck, particularly their usability. Knobs are nice.
One of the relatively popular choices for a very cheap diy scope is DSO138. It's a kit. It's basically one step above using your sound card.
Another relatively cheap option is to get a used (mediocre) analog scope. They usually cost more than DSO138, but much less than 350€.

>>1110686
Different manufacturers (TI vs. ST). There are lots of makers for 7805 and while there might be some minor differences between them, they are interchangeable as long as the package is the same. However, if there are letters between 78 and 05, then the maximum output current is different.

>>1110693
thanks for the input, much appreciated. I think I will start with a sound card which should be enough for my first projects. If I stick with it get a proper (used) one.

hiya /diy/!
electronics addict here.

pic related, a little board i'm working on for fun

glad to answer any electronics questions, been working on a lot of embedded stuff for the past 5 years

>>1110698
>pic related, a little board i'm working on for fun
What is it?
>questions
Why Allegro?

>>1110698
You sound like you're from reddit
Are you from reddit?

>>1110700
it's a temp sensor with a small lcd, the first version only did that, this is V2 where you can also record tempertures and pull the logs using USB, it uses an ST micro, stm32f042.

I use allegro because due to professional reasons, the company I work for also uses it, and while it can be hard at first, once you get the hang of it it's super powerful.

>>1110701
fuck no.

>>1110666

Do you want to be able to control the color pixel by pixel, or just one solid color that is changeable?

>>1110698
wait for the new /ohm/ thread. this ones autosaging now...

>>1110686
>Whats the difference between an LM7805 and a L7805CV?

the CV is listed by digikey as a low-dropout version. it has a dropout of 1.1 - 1.7V versus 2.0 - 2.5 for the regular 7805. diff is actually not that significant.

>>1110720
Yeah, and if you read the datasheet carefully, you'll see that even that difference is largely specmanship, as STMicro's definition for dropout is laxer.

>>1110719
>wait for the new /ohm/ thread. this ones autosaging now...

Just go to >>1102621

It only has 87 replies so it's far from dying

someone started two at about the same time

>>1110498
Post heater. I would not plug it back in before going through everything, breaker might not trip before something catches on fire. Not good for the wires in the wall either. Also, replace socket.

>>1105546
rollen

>>1102621
>>1102621
>>1102621

current thread