/diy/ - Do It Yourself

this thread's digits brought to you in part by the DRV8818 stepper motor controller/driver

Does someone know a good beginners-tier book about electronics in german? especially small-scale electronics and the like.

My language skills may be enough for posting in a frog-trading board, but i doubt it will suffice for highly technical stuff.

>>1328822
>find a nice convenient soic or qfn so i don't have to get a stencil
>it has a thermal pad
reeeee

>>1328824
qfp*

>>1328824
you won't necessarily need a stencil, but you will need flux and hot air, probably paste, and maybe a hot plate. once you have the slug attached, the rest is cake

I've got an old computer power supply. What can I do with it?

What's the common name for that pvc track they put on walls to hide cables? And also the wall plates?

>>1328891
try "raceway"

>>1328843
literally nothing

>>1328891
Check Hubble catalog.

>>1328818
I have an old repulsion AC motor, how do I ground it properly with a new cord?
There is 4 wires and you just hook the windings up to the mains.

BWAHAHA THEY ARE STILL DANCING IN THE MIDDLE

>>1328882
this source http://www.sanmina.com/pdf/resources/effectof_lead_mixing_levels.pdf doesn't find a problem with it during initial assembly, if done properly, but doing it properly can overheat sensitive components and not doing it properly can cause thermal fatigue failures in use.
this source http://publications.npl.co.uk/npl_web/pdf/matc85.pdf observes a general improvement in mixed-alloy joints and calls bullshit on the paranoia
that said, Pb-free does eat through tooling more quickly. otoh, with cheap chink tips by the bucket, it may not matter much

>>1328915
same as with the old cord
when in doubt, earth the case

>>1328934
>when in doubt, earth the case

Yeah its from the 40s so it was never grounded in the first place. Is it really easy as hooking it to the case?

>>1328938
first, check to see if any of the leads have continuity to the case. if so, try to put those on the neutral side of the power cord, then ground the case. test by plugging into a GFCI-protected outlet
in the worst case, you could install a 1:1 isolation transformer

>>1328482
>>1328882
caveat: while it's fine wrt electrical and reliability purposes, if lead contamination is a concern on the reworked assemblies, as it would be in professional work, yes you're going to want to keep two sets of tips to keep kosher. if you're not under any legal or contractual constraint to produce Pb-free boards, that little bit of contamination probably isn't going to have much effect as long as everything mixes well

>>1328824
Stick a decent heatsink on top and you should still be able to drive some power, but probably not full rated.

>>1328891
pvc cable duct

>>1328843
power stuff

What's the most visual circuit design program?
Does it allow for simulation or provide data on the fly?
I'm a very visual guy and I have absolutely shit memory.

I found vacuum fluorescent display from microwave in the dumpster near my house... And I took it. I figured out that it works. What to do now? Clock?
How many milliamps I should push into heater in typical VFD?

>>1329400
i think multisim is great but its got the ubiquitous early 2000s ui full of bugs like most professional EE software

>>1329407

As many as it takes to work.

>>1329444
It kinda worked from 18650 without any resistor, but it was glowing red. And I don't remember those thing glowing red.
And I don't have lab PSU at the moment...

What would be the best way to find VX in this circuit? I'm getting -29.1043 volts but I'm pretty certain that I'm wrong.

>>1329459
VF DISPLAY FUNDAMENTALS

>>1329463
Here's how I found vx:
I is the current in the second loop
60 = 20I - 4vx + 20I + j10I
vx = 20I
60 = (-40+J10)I which translates to 41.23 with a phase shift of -14.036 degrees
I = 1.455 ∠14.036
Vx = 29.10428 ∠14.036

>>1329400
Engineering software is hell. Think of the buggiest piece of shit you can imagine and multiply that by ten.

>>1329400
http://www.falstad.com/circuit/

Is this circuit right? I tried building it and it doesn't seem to work.

I assumed the 6-0-6 means it's a 12v center tapped transformer. Now that I think about it that seems like a mistake. The bridge rectifier will output a voltage less than half the input right? The whole point of those is to flip the bottom half of the waveform. This means the 5v LDO regulators would each be getting an input of less than 3v which doesn't work.

>>1329635
6vrms, not 6vdc. so 8.5vpk. you're getting one diode drop here because of the center tap configuration giving the regulators 7.8vpk, less as the caps discharge between cycles.

>>1329635
7805s are linear, but not low-dropout. their Vdo is 1.6-2V across rated current
what voltages are you getting? could it be a pinout difference between 7805 and the actual regulators? are your regs on different heatsinks, isolated?

>>1329660
Regulators aren't giving anything. What I'm confused about is the full diode bridge when used with a split rail like this.

The output of a bridge rectifier is less than half the input voltage right? Can someone please answer this? If the AC input gets flipped every time it goes negative the peak to peak output should be half I think.

This means either side of that diode bridge is giving 6v relative to eachother and 3v relative to the ground. 7805s need more than that to run. LDO input must be higher than output.

>>1329664
no. the rms value of a sine (eg 120V, 240V) is its peak divided by sqrt(2). so 6vac -> 8.5vac. the bridge rectifier here isn't being used as one full wave rectifier but as two pairs of common diodes, where each diode in each pair takes turns being forward biased by its half of the transformer center tap winding. so you get one diode drop unlike the standard double drop from a non-tapped transformer configuration because the return path is straight to "ground" (the tap) rather than through a second diode. i probably explained this badly but i'm too lazy to draw a diagram.

>>1329671
okay not that lazy

>8.5vac
8.5v peak

>>1329407
A bit below what makes it glow.

>>1329671
>>1329676
I get in now. Thanks. That was a huge help!

>>1329680
Okay.

Car jump start procedures often say to start the working car. What does the vehicle running have to do with its battery running in parallel with the other's?

>>1330302
A car battery is recharged by the engine and used to start the engine. If the working car isn't running you can end up depleting it's battery to the point neither car will start.

>>1330305
Do they actually have so little capacity that this is a normal concern?

>>1330313
Compared to their loads, yes.

>>1330313
No, it is rare, but batteries age.

>>1330313
Car batteries (generally) have garbage usable capacity, and starter motors pull a ton of power.

How much noise should I expect to see on a Rigol DS1054Z? I'm getting about 140mV on the cal signal.

guys i have a mosfet that is getting really hot, i also have a heatsink for it but there is no way to attach the heatsink to the mosfet. what is the best way to do this? super glue?

>>1330434
thermal adhesive

for school we need to understand a current to current amplifier but how the fuck can current flow into this opamp.

>>1330483
If it's a current-mode op-amp, they're quite different on the inside.

>>1330404
It's a Rigol dude

>>1330487
I have found people talking about this mysterious current mode op-amp but the thing is that I can't find any actual examples of a real current mode op-amp online. Does anybody happen to know the name of a current mode op-amp so I can find a datasheet?

>>1330434
Best is thermal paste + spring/clamp/bolt of some kind to keep it firmly attached.

>>1330492
TI has an app note, you should be able to find it pretty easy.

>>1330495
tnx

>>1330404

Mine looks similar, except the noise comes and goes.

I'm not sure what it means, but if you go to Acquire and change the Mode to High Res it might look a lot nicer like pic related.

>>1330483
That circuit diagram is not kosher. Maybe the presence of R1 hints at the input resistance of a normal op amp or it is of the OTA variety, but that would require a third input for the bias current and never has Ro=0.

>tfw you're at your last year of engineering master but you can't even do simple mechanics calculations
I'd kill myself but I probably wouldn't even be able to calculate from how high i should jump to ensure death

>>1330581
That's why you go into physics instead of shitty engineering.
In physics you learn how to think, in engineering you learn how not to think, but plug shit into formulas you don't understand.

>>1330599
nah i'm just dumb

I'm looking for the right motor to buy for my project (just a small DC motor, ~2.5W). I already know which values I want for rated rpm/torque, however it could be that every now and then the load on the motor might increase quite a bit.
I read that it's advisable to keep the load to less than half the stall torque, is that true? I saw a datasheet that said not to exceed the rated load.

>>1330643
Motors come with a performance diagram. At one half of the stall torque you have the max. power, which is (rpm/60)*torque. The best efficiency (for this motor) is at about one quarter of the stall torque. All this specified for operation at the nominal voltage, 5V in this case.

>>1330697
Yeah that much I know, but I'm wondering if a motor can actually get damaged if I get let it get too close to stall, or if they're all designed to withstand stall currents with no problem.
I'm indeed aiming to keep the motor in the region of max efficiency, but that won't always be possible as the load may vary and I don't have multiple gear ratios.

Since I'll be using an arduino and an l298n driver, perhaps I could set it up so that if there's too much current flowing into the motor (say, past the point of max power), then I can tell the controller to disconnect power from the motor?

>>1330707
I see no reason to waste most of the torque range. The critical condition is overheating. Maybe you can sense the motor temperature and limit it to 50°C or so, what you can touch for a second or three.

Found this guide to babby's first LiPo battery, might be a good addition to the OP?

https://www.robotshop.com/media/files/pdf/hyperion-g5-50c-3s-1100mah-lipo-battery-User-Guide.pdf

How do i become a PCB designer?

I just want to layout motherboards as a way of living

Im already an EE but i cant find a masters or specialization that allows me to train myself to do this (other than the IPC certs for manufacture)

>>1330313
It only ever takes a night of leaving the dome light on

>>1330772
that's beautiful. future anons will greatly appreciate that, if they can be bothered to read it

>>1330828
ten years ago I saw a prototype shop that had a position just for a PCB designer. you'll probably have to get close to some consultancy
>layout motherboards as a way of living
move to China

Hi there /ohm/, long time /diy/ browser but never poster, and I have a question;
Recently I picked up a motorcycle (1996 YZF600R) and the spaghetti rivals even India's electric grid.
I ask you, is there any specific name for pic related I might use to search? Can I buy them at a store or on Amazon (AliExpress has given me nothing but headaches)?

Fucking gookmoot making me buy a pass because my ISP was blocked.

>>1330918
http://uk.farnell.com/te-connectivity-amp/485043-1/terminal-open-barrel-splice-crimp/dp/1557608
you can buy them from any electronics store like digikey/mouser/allied electronics. i'm so confident of this i'm not even going to check.

>>1330918
Or go to the auto parts store, show them your phone with that picture and ask about their crimp connections.

Pic related is the motor and start relay in my 40 year old washing machine. The timer finally stopped working so I hard-wired it.

Looking at the schematic that was on the back of the washer, I deduced that the start relay was always energized, so I left it out. Well, after about 20 seconds the motor started to smoke. So I put the start relay back in, and the motor appears to be ok. I guess 40 years ago they built shit to survive idiots.

My question is, how does it disable the start winding? Does the main winding draw a lot of current while starting, and when that current drops as the motor speeds up, is the current so low that the start relay opens?

If that's the case, the motor would burn up if it ever got jammed. Years ago something did get jammed, and I messed with the basket and the agitator and got it free. But there was no smoke. How does the start winding survive that? And that thing I forgot to label beside the start winding is labeled O.L., which must mean overload. It must be a current sensor and not a thermal sensor, base on my smoke test.

Incidentally, this is the configuration for spin mode. In agitate mode the main winding is reversed, but everything else is more or less the same.

Is there a risk in having two motor encoders very close to each other? Like if I have two motors, each with an encoder on the back side, and the encoders are facing each other with a 1-2 mm gap between them. Can they interfere with each other's readings?

>>1331152

The only kind I can think of that might be a problem is magnetic, and even then, only one specific type of magnetic encoder.

If optical (which is what I'm assuming), then, no, you shouldn't have problems.

>>1331152

that looks like a magnetic encoder. I'd be more concerned about the two motors fighting each other, but in reality I doubt if there would be any interference.

If you have room you could put a sheet of ferromagnetic material (most steels), and while that will shield the two from each other, there could be some hysteresis or eddy current drag that will load your motor.

If you are getting the 48 cpr or the 64 cpr version, the magnetic poles are so small I doubt there will be any problem.

What's be a good way to filter or regulate the 5V supply from a generic phone style USB plug.
I want to drive some nixie tubes so I'm going to have a 5V to 170V converter.
However, I'm also going to use the 5V for my microcontroller, RTC, a gps module, some shift registers and binary to decimal converters.
I know the output from those USB plugs can be fairly noisy and have a bit of ripple so would it be okay to just use an LC filter or should I use a regulator?

>>1331166
>should I use a regulator?

regulators tend to need some overhead, meaning they can't regulate 5v to 5v.

why not a large electrolytic cap?

>>1331169
Yeah, I was thinking I'd need closer to 6V to get 5V out of the regulator.
Maybe I'll use a 9V supply instead and feed that into a 5V regulator

>>1331159
>>1331161
Yea sorry i didn't link the actual thing: https://www.robotshop.com/eu/en/pololu-12v-2041-metal-gear-motor-mp-48-cpr-encoder.html

It's hall effect i think, 48 cpr.

>>1331193
My choice for minimal distance would be the thickness of the magnet wheel.

>>1331011
The relay is an SPDT relay, in the normally-closed arrangement. Before being turned on current can flow through both the start and main windings, but once the current through the main winding increases enough after a small delay (since V = L*dI/dt), there's enough current passing through the relay's coil and the start winding is deactivated.

The start winding isn't meant to be run for more than a fraction of a second, and is probably made of pretty fine wire with a relatively high resistivity so it heats up if run for too long. I didn't know a starter winding could be used without a starter capacitor, but for something which disables within the first cycle I suppose it makes some sense.

>>1331245
I'm gonna have to buy a different case then, fuck...
Unless I just make a hole on the side and let the motors stick out a bit

>>1331249
Let the experiment decide. There's also the option of a 0.5mm soft steel shield should there be any interference.

>>1331253
That would have a higher magnetic permeability than the air and make the parasitic magnetic field on the other end much stronger. What you want instead is a superconductor or some other significantly strong diamagnetic solid.

Faraday cages are in no way blocking magnetic fields alone, but they block the combination of the electric and magnetic fields. As far as I know, the electric field's alternation causes a voltage between two points on the cage, which cancels out the electric field within it. Since the electric field component of the rotating magnet wheel is stupidly small, a faraday cage or similar enclosure will have no effect.

>>1331256
Looks like you can buy a 25x25mm square of 1mm thick bismuth foil online. Would that do?

I am going though the book "Digital Design by M. Morris Mano and got to the practice problems at the end of chapter 1. Do I really need to know how to add and multiply numbers in different bases?

I am currently tying to evaluate this problem: (135.4)_4 x (43.2)_4 and holy fuck I have never had more difficulty with math than I am having right now. The problem instructs not to convert to decimal so I am forced to do it with the base in mind. I have literally spent 30 minutes on this at this point and I am no where near close to getting it right. Is this something I will need to be able to do when designing logic circuits? This seems absolutely useless.

>>1331275
trying*

>>1331263
Bismuth has a relative permeability of 0.99983363 that of air, so the effect will be less than negligible.

>>1331275
>Do I really need to know how to add and multiply numbers in different bases?
No. At most you'll have to switch between decimal, binary and hex when programming digital stuff, for example, when using port manipulation and masks, but any modern calculator can do the job 4u. I've never encountered a problem irl where I needed to multiply numbers in other bases.

It's not too hard anyway, just multiply normally but only carry a one at 8 or F or whatever, instead of 10.

>>1331284
i was just going off your mention of diamagnetic solids, of which bismuth seems to be the strongest natural occurring element. But maybe I totally misunderstood

>>1331166
>generic phone style USB plug
do you even have enough power from the USB adapter to
>drive some nixie tubes
? voltage boosting is more lossy the higher the boost factor. design and test this first
>5V for my rgbhlrhlrlrb
>not using 3.3V
take a good, hard look at your life, son

>>1331263
>he hasn't heard of mu metal

>>1331275
if you intend to design ALUs, yes
otherwise, not really

If i have a circuit with 40 transistors and another with 20, which would be faster?

>>1331321
Assuming they are all the same size btw

>>1331293
It is the most diamagnetic naturally occurring element, but it's still far away from being useful for blocking magnetic fields.

>>1331316
>mu metal
Maybe if you want to conduct the magnetic field literally 100,000 times greater than the air, but that is the opposite of what we're trying to accomplish here.

>>1331321
Depends on the context. If you're talking about propagation delay with logic gates, and the transistor logic gates are feeding one another, then 40 gates will take more time. But if you're doing parallel operations (a bit of a stretch with only 20 transistors) then 40 will take less time to do two parallel operations than 20 doing those operations one after another. But the first answer is probably what you're looking for.

>>1331316
>>1331275
>>1331287

Thanks, and I do intend to go into computer design (a long-standing goal of mine has been to replicate the IBM 1401 in 7400 series logic). In any case, after thinking over it for a long time, I found a simpler way of doing multiplication that works for any base-r number:

Let's say you want to multiply 3 and 3 (both in base 4). You take the product of 3*3 (9) ad divide it by r (in this case, 4). This gives you 2.25. The whole number here is the number in the 4^1 place (as I understand it), so the product in base 4 is 2x. Yu then find x by multiplying r by 2 (or haterver number you got for that place) and subtracting it from the original product. In this case, this would be 2*4 = 8; 9-8 = 1. 1 is the number that goes in the r^0 place. So, the product in base 4 would be 21.

If you have a carry digit, you can just add it to the initial product and go though these steps and it works fine.


For addition, it is as simple as adding up the two numbers in question, dividing by r (to get the same number in the r^1 place) then multiplying that number by r and subtracting the product from the initial sum to get the number in the r^0 place. So 2 + 3 = 5; 5/4 = 1.25; 1*4 = 4; 5-4 =1; sum in base 4 = 11.

Sorry if I sound like a brainlet, but I'm happy that I could do it without resorting to converting to decimal.

>>1331321
it depends on the length of the critical path. for a practical example, compare an n-bit adder with synchronous carry vs. ripple carry

>>1331333
>computer design
>74xx
you'd do better to learn higher-level principles, I think. a superscalar 1401 clone in verilog could be fun :^)
>replicate the 1401 in 7400
damn son, that's a lot of chips
>multiplication
ah but you kinda did convert it to a base 10 decimal, which isn't halal. since the distributive property of multiplication applies regardless of base, you can use the same algorithm you learned in grade school, given the correct multiplication table. for base 4:

0 1 2 3
0 0 0 0 0
1 0 1 2 3
2 0 2 10 12
3 0 3 12 21

just as 13 * 22 = 2 * 3 + 2 * 10 + 20 * 3 + 20 * 10 (all base 10),
13(4) * 22(4) = 2 * 3 + 2 * 10 + 20 * 3 + 20 * 10 (all base 4), so
12 + 20 = 32
+ 120 + 200 = 320
so, all in base 4,
32
+320
----
11 (carry)
312
----
1012(4)

just to check, we'll convert to decimal: 13(4) = 7(10) and 22(4) = 10 which * = 70(10)
1012(4) = 1 * 64 = 64 + 4 * 1 = 68 + 2 = 70
fitting digits, btw. don't worry, we serve all* kinds here
*except tripfags, namefags, and most shitposters

>>1330828
Layout-only is considered grunt work. We have a layout department at work and they are mostly contractors.

You're better off being a designer who is also capable of laying out boards. Pick it up yourself by studying commercial boards and learning the lingo and process.

I am wanting to set up a solar and wind system yo help offset the utility cost or completely sustain it.
I use about 2.5-3KW a month, so I doubt I'd ever completely sustain my energy needs with my limited space(about 200 sq ft of space I can have for panels alone). Anyways. Why I am here is because I don't know the best way to go about this. What kind of battery do I need? If the solar kit I'm looking at is rated for 24v, does that mean I should never add batteries that go over or under 24v? Is that why I see so many 350ah batteries rated at 6vs so you can stack them? Sorry if I am a total dumb ass about this.

>>1331253
>soft steel
That's how microphones, audio transformers, loudspeakers and some mains transformers are magnetically shielded. It is known to work.

>>1331454
Shielded from EM radiation, not from slowly changing magnetic fields. Putting a compass in a (non ferromagnetic / large enough) faraday cage doesn't prevent it from showing which way North is. Needless to say, putting it in a small ferromagnetic cage will just cause the needle to be attracted to the cage itself.

>>1331333
>divide it by r (in this case, 4). This gives you 2.25

You're converting to base 10 there, 9/4 in base 4 = 2.1

There's no need to do division anyway, you're only multiplying. Just follow standard long multiplication.

>>1331417
>2.5-3KW a month
You are confusing power and energy. There are many tutorials about planning solar and wind energy systems that offer basic understanding.

>>1331462
How would you solve the original problem?

>>1331479
See if it works without any modifications. If not, change the geometry.

>>1331248
>The relay is an SPDT relay, in the normally-closed arrangement.

thanks for the reply, but it clearly cannot be a double-throw relay as there are only two wires going to that side.

And it's not normally closed. It uses gravity to hold it open (has an arrow pointing to TOP), and if you flip it over it makes contact, but in the normal orientation it is open until power is applied to the coil.

Anyway, thanks.

>>1331462
>not from slowly changing magnetic fields. Putting a compass in a (non ferromagnetic / large enough) faraday cage

a sheet of ferromagnetic material is not a farraday cage, and will shield one small component from magnetic fields coming from another small magnet on the opposite side, whether moving or stationary, provided that the shield is not magnetically saturated, which a tiny disk with 24 poles cannot possibly do.

Hello friends.

I have a problem I was wondering if you gentlesirs could help me with.
I have a Siemens S7-300 F model PLC, i.e. a failsafe version of the industry standard.
On it, I have a signal for a load cell that is no longer in use, that is troubling me a bit. My crane is working fine, but even if I have made the PLC allow wire breaks without going into failsafe mode and shutting everything down, it's still searching for the missing signal, and is giving me a signal fault in the module, and resulting ugly red light.

My customer doesn't want that red light, and neither do I, but I can deactivate the channel, as on the F-CPUs you need to activate channels two at a time, and I can't make that happen.

So, long story short, I want to install a resistor on the zener barrier going to that channel to simulate a signal. So for a 24V 4-20mA signal, am I right in thinking this can be done as simple as 24V / 0.01A = a resistor of 2400 ohms to simulate a 10mA signal, or will this fuck me over?

>>1331590
**I can't deactivate the channel

The batteries from alibaba are a lot cheaper, am I going to die if I order one from china or do all batteries come from there anyways?

>>1331633
Sorry, should have specified I'm looking at the huge ones for home generators.

>>1331590
Isn't a load cell a 4-wire device?

>>1331640
It's got an internal amplifier, so it only uses 2 wires.
Plus from transformer, and signal back to PLC.

possibly (probably) a stupid question but can I connect the output of 2 charge pump circuits together.
For instance if I had a charge pump circuit for 100v that could deliver 50mA to a load but I needed 80mA, could I just combine 2 of them together?

>>1331642
When you connect a 2K4 resistor between +24V and ground (!) you get a current of 10mA. Do you _know_ that the signal input of the controller expects a current and not a voltage? In the latter case you would feed it 24V. I would at least use a divider and calculate the resistors such that the input sees both a compatible voltage and current.

>>1331688
With normal diode multipliers you can do that.

>>1331695
>When you connect a 2K4 resistor between +24V and ground (!) you get a current of 10mA
Yeah, but I think connecting it to plus and signal will make the connection to ground through the signal receiver. No grounds are connected straight to components afaik.

I'm pretty sure it's reading the milliampere current and not voltage.
The problem is I'm offshore on an oil rig atm, and as soon as the wind and swell goes back down I'll fly home, hopefully to never return, and just have the rig electricians install this after I mail it to them.
So it should ideally work straight out of the box. Thanks for responding.

Pic related is my module; https://mall.industry.siemens.com/mall/en/ww/Catalog/Product/6ES7336-4GE00-0AB0

>>1331703
AC isn't an option unfortunately

>>1331742
Then combine the two outputs with two diodes.

>>1331688
If they're normal unregulated charge pumps, usually you can. Run them 180 degrees out of phase if possible.

I'm already an EE, I really like laying out PCBs I can spend days doing it, measuring components, reading datasheets, however after looking for jobs for a while to do this and i haven't found a single one, to do layout only

PCB Design seems to be a full specialization, however companies expect you to do the programming, product design, industrial design, design pcb, high and low level programming... which i am capable but... I would like to just do PCB layout is there such a job about this?

would getting certified on IPC-A-610, IPC-J-STD-001, IPC-7711\IPC-7721 help my get me a job?

Any suggestions of a program, masters, course i can do to just to live by doing this?

>>1328818
Sorry if this is the wrong place to ask, but I have a degree in Biology and am looking to go back to school for EE. I have been reading the beginner books listed and working with my kit, is there anything else I should be learning?

>>1331404

even high speed pcb design?

>>1331804
build a lab and always put in practice what you learn

https://www.youtube.com/watch?v=R_PbjbRaO2E

I like hackaday com to

>>1331366
Thank you, I appreciate that anon. Also, just for my own clarification, this area of study is best described as computer engineering, correct? Or would it be a specialized area of electrical engineering?

>>1331695
>he doesn't know about current loop signalling

>>1331704
that should work to fake a 40% reading

>>1331879
no problem, sorry the leading spaces got missed and it didn't turn out as readable as it should have
>computer engineering
you is correct. the discipline is about 70% computer science, 70% electronic engineering
Hennessy and Patterson's books "Computer Architecture: A Quantitative Approach" and "Computer Organization and Design" are the standards in the field. great winter reading

>>1331804
>>1331810
also
>https://www.youtube.com/watch?v=R_PbjbRaO2E
don't do drugs

>>1331888
>"Computer Architecture: A Quantitative Approach" and "Computer Organization and Design"

Thanks, these look really good. I'll definitely pick them up and give them a read.

>>1331537
Well that's strange.

>>1331539
When I put my high-permeability transformer core in between two magnets, the magnetic attraction between the two magnets gets stronger, i.e. it conducts magnetic fields much better instead of shielding against them. Regardless of what happens when the magnetic poles are moving (hysteresis, etc.), this is not a desirable effect as the encoders need to work when stationary. In no way will having a high-permeability material on one side of any collection of dipoles will cause the magnetic field on the other side to be any weaker, provided the high-permeability substance is uniform. Perhaps a sheet of thick copper would be able to average the magnetic field on the other side of a multipole magnetic ring via its hysteresis, but when stationary this is useless.

The faraday cage stuff was assuming that those suggesting the ferromagnetic material for shielding were thinking along the lines of a faraday cage.

Also this helped me find out that those magnets I tried (and failed) to use to make an electric motor in primary school didn't have a pole at each end but on either side, which also explains why my bootleg guitar pickup made with that same magnet only picks up 50Hz. Progress!

>>1331965
Conducting magnetic fields is exactly what you want to shield a magnetic sensor, you concentrate the flux through the shielding medium parallel to the sheet to reduce the flux going through it perpendicular to the sheet.

Look at all the materials advertised for magnetic shielding, Mu-metal, permalloy for example, they're all very high permeability.

>>1331888
>don't do drugs
What do you mean with this?

Also, digits!

>>1332056
If the sheet is homogenous, it should increase parallel magnetic fields just as much as perpendicular ones. It would just act similar to a substance with a higher optical density if the magnetic field were replaced with light.

But I think I get what you're saying:
>the Hall-effect sensor is to the side of the magnetic ring, not underneath or above it, so it measures the magnetic field coming radially from the ring
>having another magnetic ring coaxial with the first one will produce a series of "field lines" that first come out more axially than radially, but then curve around to the side and become more radial at larger distances
>by putting a material of high permeability in between the rings, the flux travels less effective distance and therefore has a significantly smaller radial component

This does (probably) rely on the two assumptions that:
>the flux is emitted axially
>the sensor is mounted horizontally
which appear a little counter-intuitive, since you'd want the magnetic poles pointing at (and away) from the sensor. If the flux is instead emitted radially, then it moves LESS radially at a larger distance (or lower permeability),
It looks like the the sensor is mounted horizontally (>>1331245) but I think the magnetic poles must also be radial.

If I got something wrong here then please tell, but for sure a superconductor (or other relative permeability << 1 substance) would shield it perfectly (albeit introducing some magnetic friction), right?

>>1332071
This is what I was thinking:

>>1332071
Pretty sure it is radial flux as you described. Let's simulate it.

Super simple setup, a slice through one side of radial magnet, points representing where both sensors would go, and a box that can represent air or a shield material.

>>1332095
Flux lines and density plot with no shielding material

Then with near superconductor ur = 0.001 like you wanted. Shields great, very little flux at the sensor point on the other side, but not particularly practical.

and finally with some kinda shitty ferrite ur=250, also near perfectly shields the sensor on the other side, just by almost completely concentrating the flux inside it compared to the superconductor almost completely repelling the field, both approaches work for shielding.

>>1332102
Aha, so that's how fields react in high permeability solids! I figured the field would be elongated, not contracted. But I guess it would have elongated the field if it were axial? Now time to graph magnetic flux density on the other side of the barrier as a function of magnetic permeability...

Thanks for clearing that up, and that's a pretty damn neat simulator too.

For reference to servo anon:
>>1332076
"magnetic flux rings.png" is wrong in the conclusions it makes about permeability, the magnetic simulations are correct. You should probably go for some electrical steel laminations from a transformer or something, but other sources of magnetic iron will work fine. If you're going to buy something instead of sifting though the trash heap, then (half) a flat cable ferrite might work if you can't find an appropriate ferrite disc.

>>1332117
The simulator is FEMM http://www.femm.info/wiki/HomePage It's free and pretty easy to get started.

Set up for axial flux it does spread it out over a wider area, but sill has the shielding effect.

>>1332060
I was merely
>implying
that Dave Jones is a meth head, all in jest of course

>>1332056
You can indeed redirect the flux. Do this at home:
https://www.exploratorium.edu/snacks/magnetic-shielding

I have a SIC Marking e8 machine that keeps blowing its fuses. I don't do a lot of electronics but have a pretty good electrical understaning. When I checked the 240v side there is no short between either the active or neutral or to earth. I have attached a pic of the pcb where I have already removed the fuses and a cap that looked dodge but is not the fault. Wondering if any of you wizards can help me find the fault or point me in the right direction. Cheers

>>1332172
are you using the correct fuse speed (fast vs. slow)? fast-blow fuses may blow due to the inrush current that is fairly normal for a switch-mode power supply
are you sure it's not the marking head that's gone bad?
if you do not have a variac and a scope, you are not equipped to repair a switch-mode power supply. refer servicing to qualified service personnel

>>1332129
Nah, just an over-the-top personality. Strangely enough, he's the only one where I like that.

>>1332232
it's like Billy Mays and Yo Gabba Gabba had a love child and I just can't hang with it
I managed to make it to the end of the episode on the reflow toaster oven controller and that was at least half wasted. not a fan of the vlog format

>>1331152
This tutorial (http://andrewjkramer.net/motor-encoders-arduino/)) says:
"It’s also important when installing your motors to not mount them too close to each other. If your motors are mounted back-to-back, the magnetic encoder wheels will interfere with each other. I initially had only about 5mm between the two encoder wheels. The magnets in the encoder wheels linked with each other and made it impossible for the two motors to turn at different speeds. Keeping the encoder wheels at least 20mm apart seems to avoid any interference problems."

>>1332261
Different encoder, axial field, the other is radial.

>>1332273
i didn't know there were different types. Is the difference between the two exactly what it sounds like?

>>1328818
is it ok to hit the limits with the output of an opamp as long as I know about it or is it a thing I should prevent from happening at all?

>>1332290
>is it ok to hit the limits with the output of an opamp as long as I know about it or is it a thing I should prevent from happening at all?
you mean clipping?

won't hurt the amp, if it's something you want (adding harmonics, etc.) go right ahead

>>1332292
>you mean clipping?
yes, didn't know how it's called. Thanks

>>1332288
Yes, Ctrl-F hall-encoder.jpg
also, topic shielding

>>1329502
I miss my circuit classes now

>>1331748
>>1331750
I'll be using 555 timers for the charge pump.
I'll have an output voltage of about 190V.
Would two 1N4007 diodes be suitable to connect the outputs together?

>>1332368
>timers for the charge pump
There may be a nomenclature mismatch..
A circuit diagram would be useful.
>two 1N4007 diodes
I'd say yes, even two 1N4004 diodes.

>>1332333
What did you enjoy about them?

>>1332412
not him but it was like doing sudokus that actually help you land a job

>>1332415
>help you land a job
Employer says too much analysis, not enough synthesis (creative design) = too many unemployables.

Trying to modify an old uhual trailer into a living space. I plan on powering two chest freezers, a small crt, laptop, chargers for electronics so far and also want some elbow room for the future. Am I fucking up?

>>1332124
Eh, FEMM isn't for my operating system.

So how does this fit with my anecdotal evidence of:
> When I put my high-permeability transformer core in between two magnets, the magnetic attraction between the two magnets gets stronger (assuming the magnets are the same distance apart in both situations)?

>>1328818
Now that Radioshack is dead, what PHYSICAL store can i go to if I need an emergency cap or resistor or someshit. Mouser is cool and all but I dont want to wait 5 days for a 10cent cap

>>1332424
employee says too little pay for patent licensing = too many shitty employers

>>1332290
I think we call those rails?
touching is fine but be careful going beyond them. many ICs will misbehave when their pins are taken outside the Vcc-GND range. some op amps will not provide gain or perhaps even misbehave if you take the inputs beyond the specified range, which may be smaller than Vcc-GND (e.g. LM324-type amps are only specified for inputs from GND to Vcc-1.5 or so)

>>1332551
there's no single answer. you might have an independent electronics retailer near you. if not, you might try your local stereo/TV/amplifier repair shop but you might pay a buck or two for that cap
better to prevent the problem and order an assortment from the chinks after the New Year

>>1332551
there's no more franchises so you just have to hunt around your area on google. south florida has MPJA, and even as a dedicated hobbyist shop it doesn't sell very many passives.

>>1332404
Yeah, a 555 timer can be used as a charge pump apparently.
Looking at it now though, it's actually set up as a switching boost converter in the circuit I'm planning to use

>>1332484
FEMM works fine under Wine. Or use OctaveFEMM.

> the magnetic attraction between the two magnets gets stronger

You sure the change in force is caused by increased attraction between the magnets, not just the additional attraction of both to the ferrite?

>>1332685
>wine

Well it works with a magnet on one side and a piece of iron on the other, like sticking a magnet to your screwdriver so you can pick up nuts.

>>1332687
Shape matters. Magnet on screwdriver works nicely since it concentrates flux down through the tip Stick the same magnet on a plate and you'll significantly reduce the picking up power.

>>1332672
Charge pumps have switched capacitors, not a switched inductor like your circuit. This is a typical step-up converter used for Nixie tubes.

Running regulated power supplies in parallel to increase output power can be tricky. I would first try to use 12V as supply instead of 9V and use an inductor that can handle more current without saturating.

If I have a 0 to +30v single output power supply, and want to build a circuit that requires -10v and +10v, is there a tricky way I can power the circuit without another power supply?

>>1332861
Basically yes.
Set your single supply voltage to 20V and use a power opamp to buffer 1/2 the supply voltage. This provides a steady voltage on the output that can be used as a virtual ground reference.

>>1328891
wiremold

>>1332861
>>1332871
Beat me to it
For low current applications, you can also use two resistors to make a virtual ground.

Is there a way to power a rpi and arduino with a single battery (and only one voltage converter)? It seems that arduinos want 6-12V while the pi only takes 5V. I have a 3S LiPo so i'm kinda tempted to just feed the arduino directly but apparently at full charge the battery could measure as much as 12.6V, so the arduino would be at risk of heating up too much.
Either way I'm also thinking of having a small fan in the case where i'm putting all this shit but idk if it'll be enough, and besides i'd be wasting a lot of energy.
Right now my solution is to just have two buck converters but i was hoping to find a way that takes up less space and money.

>>1332871
>>1332880
Thanks

>>1332892
https://electronics.stackexchange.com/questions/65576/arduino-uno-r3-directly-supply-regulated-5v-to-5v-pin

short answer, connect 5V to the USB port's 5V pin.

Hello, any recommendations or protips on a projected test setup for study, enjoyment and demonstration of faraday’s laws of induction (pic related)
I am wondering how to calculate the wire gauge, coil size, magnet size, number of loops in the coil, and how to combine these and produce repeatable results with digital multimeter which can meter millivolts and microamperes.
I am aiming for a upright tubular construct in which a magnet is dropped and because it is drawn by gravity through the coil, it produces a steady repeatable velocity for magnet flux change.
But it needs to be viewable in a normal multimeter, and perhaps drive a small load, a small light bulb, or if that is not applicable, a LED?

Hey /ohm/, I'm powering an Arduino nano with a 3.6 V Li-Ion battery running through a DC-DC converter to 5V. I'm running some 5V solenoids, and an accelerometer on here as well.

Disconnected from the USB, I can set up a simple loop to toggle the solenoids at their rated voltage of 5V. Connected to the USB, and disconnected from the battery, I can run the accelerometer just fine. However, if I connect both my external power and USB to the Arduino, which power supply will it use? If I power using the USB, then I can still actuate the solenoids but they get really hot since they need to draw more current when running at 3.3V.

Is there a way to use the USB exclusively for serial communication, and not power?

>>1332996
Right, I forgot to mention why I want this. I need to be able to view the accelerometer outputs for debugging, and also actuate the solenoids. For example, I want to energize the solenoids when a certain acceleration threshold is reached.

>>1332996
>>1332998
the usb supply is fed through a diode. it'll generally be using power from the battery, unless that voltage drops out in which case it'll draw power from usb.

disclaimer: this is based on an "arduino nano schematic" google search. chinese models may vary.

i experimented with press fitting pin headers (.64mm square into .8mm hole). it worked, but i wouldn't recommend it with more than ~3 pins at a time.

>>1333025
Thanks!

Can I measure AC amps by putting my meter in series with the load?

>>1333196
yes as long as you don't exceed the fuse rating.

>>1333196
>>1333201
And assuming your meter has an AC amps function, under the assumption that someone who owns a Fluke 87V would not be a person who doesn't know how to use the AC current range. A DC current range will not work.

>>1333196
No need to put in series.

>>1333315
Or a fucking $2 current transformer.

>oh no, smashed a workbench full of relays
>sell as surface mount

>>1333317
>$2 current transformer
Not U-shaped, unusable for daily work.

>>1333346
no u

What exactly is "UNGRD"?
I hooked up the Neutral white wire to post 1 UNGRD
Hot Black wire to the post 4. This is how the old wires were hooked up, but it had a non polarized plug and no ground

I added a power switch that interrupts the hot black wire.

I wired the ground wire to a screw on the motor.

Does this seem ok? I mean it works fine, but im afraid of shocking myself or something. Im no electrician, so im crash coursing on google...

>>1333468
Also its just an old regular 110v AC motor, USA

>>1328818
This is probably not a good place to ask this, but here it goes
I have a couple of ESP8266-01s, and I am trying to implement I2C bus by bitbanging it. Now the problem is they have only 2 GPIOs, excluding UART, my soldering skills are not good enough to directly solder a wire on the I2C pins, so I programmed the GPIO pins to act as SDA and SCL. I am pretty sure the SDA and SCL lines are normally high and are driven low for '0's, meaning they are open collectors. My GPIO pins are set as input
I am using the ESP8266's internal pullup resistors to drive the lines, by driving line 'high' and 'low' states by toggling the internal pullup resistors and drain respectively but its not working.

Where are the points where i should checking for mistakes?

>>1333317
i wanna see a $2 current transformer
educate me senpai

>>1328818 (OP)
This is probably not a good place to ask this, but here it goes
I have a couple of ESP8266-01s, and I am trying to implement I2C bus by bitbanging it. Now the problem is they have only 2 GPIOs, excluding UART, my soldering skills are not good enough to directly solder a wire on the I2C pins, so I programmed the GPIO pins to act as SDA and SCL. I am pretty sure the SDA and SCL lines are normally high and are driven low for '0's, meaning they are open collectors. My GPIO pins are set as input
I am using the ESP8266's internal pullup resistors to drive the lines, by driving line 'high' and 'low' states by toggling the internal pullup resistors and drain respectively but its not working.
Where are the points where i should checking for mistakes?
I am using a pcf8574 I2C to serial port expander with on the other end if that matters. There are no other devices on the bus

>>1333201
>>1333305
Cool. I had read something about not putting meters in series for AC stuff, but I figured it was probably just because clamp meters are so cheap for AC that you shouldn't bother.
>>1333315
I've got a couple of clamp meters, I was just curious because of what I had heard.

I don't use my meters enough to justify the level of equipment I have access too, but I won't look a gift horse in the mouth. The 87V with the calibration certificate is my dads, as well as the 323, and the 179 is my friends, the others are mine. UT216C was a gift, 87 got passed to me when my dad got the calibrated 87V, and I bought my 87V for $86.

>>1333505
Got an oscilloscope? Have you read the pcf8574 data sheet for all the important parts? Is your i2c address correct?
Do you have a known-working implementation of i2c to compare your waveforms to?
You could even implement it with your other esp device configured as a slave just to make sure the most rudimentary communication works.

>>1333505
nah, this is a fine place to ask this sort of thing
what documentation are you working from? NXP has a lot of docs on the nuts and bolts of the I2C protocol in an app note somewhere. I'd start by reading that, especially timing info
might try leaving the pullups on and just setting the pin direction in vs. out to drive the line to 0
maybe, for giggles, you should multiply all your delays by 100000 and hook up LEDs and switches so that you can manually monitor and interact with the interesting points of the protocol, making sure stop and start conditions are properly timed, clock stretching and ACK turnaround works, etc

>>1333589
I don't have an oscilloscope but that's good advice there, I do have a couple of AVRs lying around, I could use them to debug the bus.

>>1333597
Afaik i2c is extremely tolerant of the clock to the extent one could drive an i2c bus by pressing a couple of push buttons, but I have no idea how true is that in practice
I'll try messing around with the pull ups as you suggested

>>1333497
https://www.aliexpress.com/item/2pcs-NEW-0-12-300-turn-Current-transformer-leakage-magnetic-coil/32648964952.html
2 for $1.50, or just wind your own. But the $7 or so CTs with a split core are more useful for current measurement.

>>1333611
anyway, also try adding real 4k7 pullups, the ones built-in to the ESP8266 may be insufficient to overcome the capacitance on the bus in time to meet rise time specs
https://www.nxp.com/docs/en/user-guide/UM10204.pdf

>>1332483
Keep in mind that wattage ratings for inverters are for resistive loads. Your chest freezers have induction motors and they will draw some reactive power. if you only consider the nameplate horsepower when sizing your inverter, you're going to be unpleasantly surprised.

what's the CRT for? If you're living in a trailer off solar power you should optimize everything for lowest power consumption. I don't think there's any reason in this decade to hold on to CRTs.
On that note, you should doubly insulate your freezers so they don't run so often. You probably will never have as much power as you think you will and certainly not as much as you'd want.

am i stupid for wanting to learn pic assembler? should i just use C and save myself some pain? i have plenty of desktop programming experience but i've only done verilog and arduino C for hardware.

>>1333723
I'm doing something similar with a van, and I'm running into no end of issues trying to figure out what amount of solar I want and how much battery I'm going to need. It doesn't help that everything seems to want to use a different measurement. I'm looking at 100W 12V DC solar panels, batteries that say they are 235Ah 6V DC, and I have no idea what my draw is for a RV minifridge since everything online says it will be "about $47 a year" which doesn't tell me a goddamn anything.

By any chance do you know of a listing somewhere online as to what the average power draw from things are? Knowing how much draw I can expect from just the fridge and AC alone would make this a lot easier.

>>1333747
C is more practical. If you only have enough time to learn one, go with C.
Assembly is entirely appropriate though for tight loops and tight ISRs and if you need timing accurate to single instruction cycles. Also if you need deterministic execution that isn't subject to compiler decisions.
PIC assembly is kind of shit though.

can anyone recommend me a cheap linear power supply? I need to get 3.3V, 5V, and 15V outputs. Can I do it myself with linear regulators and a transformer?

>>1333747
If your goal is experience with learning any assembly language, go nuts. It's a bit weirder than doing AVR assembly for example since you've got to deal with the single working register, but it's still a pretty simple instruction set to get started on.

For a practical application, unless you're making thousands of the thing you're programming, it's not worth your time thinking harder to get your task to work in assembly when you could just get it done in C instead, possibly spending the little bit extra to get a one step faster/larger chip if necessary to deal with the bit of compiler overhead.

Writing C for a PIC is pretty similar to writing C for AVR/Arduino.

>>1334001
>Can I do it myself with linear regulators and a transformer?
You can.

> I need to get 3.3V, 5V, and 15V outputs.
How much current on each? Why linear?

>>1333914
Find what value they're using for power price, plug and chug http://www.wolframalpha.com/input/?i=($47%2Fyear)+%2F+($0.12%2FkWh+)

If it's the US, around 12c/kWh is average, but varies quite a bit by state.

For battery capacity, just multiply amp hours by nominal voltage for a decent approximation of energy. http://www.wolframalpha.com/input/?i=235A+h+*+6V+in+kWh

For the solar panel, note the 1000W figure is best possible case, there are various calculators online to get a better estimate of how much energy you'll actually get for a given location and how you're angling the panel.

>>1334004
I'm sorry, I couldn't tell you exactly what current, but I can give you an idea as to what I will be using it for. 3.3 and 5V will power an arduino and both a DAC and an ADC. The 15V will drive a piezoceramic actuator (which requires the most precision)

>>1334007
Got specs or a datasheet for the actuator?

Unless it's a really high performance DAC/ADC I'd assume something like 100mA max for the 3v3 and 5v, so just a linear regulator off the 15V may be good enough if you don't mind poor efficiency.

>>1334008
https://www.murata.com/~/media/webrenewal/support/library/catalog/products/sound/p37e.ashx
page 21, 9kHz buzzer.

The DAC and ADC are both 16 bit. If those datasheets are necessary, here they are:
DAC
http://www.ti.com/lit/ds/symlink/dac8814.pdf
ADC
http://cds.linear.com/docs/en/datasheet/232616fa.pdf

>>1334017
nice dac

>>1331249
You might be able to rip some mu-metal from a hard drive and stick it between the encoders.

>>1334006
it appears anon wants peak, not average

>>1334007
you have only stated what you will be supplying, not what you will be using it for. if you're so concerned about noise you will probably need to decouple the lolduino heavily anyway, and it goes without saying that high-precision data converters will need careful enough attention to supply decoupling that input noise would hardly matter
tl;dr: lern2inductor

>>1328818
I don't get the joke in the picture plz help

>>1334060
The previous thread had some guy asking how to understand capacitors, because he didn't know how they worked.

>>1334062
Also kinks in his leads = sexual kinks

Would anyone be able to recommend a Qi wireless power Tx/Rx pair?

At the moment I want to try using NXQ1TXH5101J (Tx) & BQ51013BRHLT (Rx) but just want to see if there are better suggestions out there ;^)

Fast question. This cap is 1uF or 1nF? It’s pretty old.

>>1334172
1000p=1n

>>1333914
>I'm looking at 100W 12V DC solar panels, batteries that say they are 235Ah 6V DC,

A little background on watts and amp-hours.
Batteries store charge. Electric current is a flow of charge. The standard unit of charge is the coulomb, one amp flowing for one second moves one coulomb of charge. An amp-hour is just a larger unit of the same thing.
Voltage doesn't factor into that at all. The voltage of a battery depends on its chemistry and varies with how charged the battery is, but it doesn't vary linearly.
To figure out how much energy (in joules, or watt-hours, whatever) you have to integrate over the voltage. You can get an approximation by assuming the voltage stays constant, and just multiplying by the voltage.
But that's why batteries are rated in amp-hours, because fundamentally they store charge. Some laptop batteries do have ratings in watt-hours but they rate that by assuming some typical voltage discharge curve. If you overcharge or undercharge even by fractions of a volt, the energy stored will vary a lot.

>>1334007
also consider using a switcher for everything but the DAC/ADC themselves which get their own linear regulators. no problem at all as long as the ground is common

>>1334180
Okay, this is the last. Is it 5,6nF or pF? It’s from an old tube oscilloscope.

>>1334228
5p6

If you want to get into pure hardcore electronics, like DLD and Signal Processing, then go to a university library.
If you just want it as a hobby, and want to make robotics and stuff, then don't waste your time on this. Buy an open source controller like ARDUINO, and start learning the programming basics behind it.You can find tones of material on it

>>1333468
>Does this seem ok?
Yes, it sounds fine.

>>1334268
>Buy an open source controller like ARDUINO
t. guglielmo venedetti

>>1328818
I could really use some help.
Some time ago, I build a habitat for my frogs. I put some super bright blue LEDs behind some pieces of geode with the intent to light them. I wired everything properly and they all did light up when I tested them. My intention was to put a switch on them so I could just turn on and off, but life got in the way and I never attached the switch. I would like to finish my project after two years, but I don't remember which wire is which.
Is there a chance to damage the LEDs is I use a 1.5v or 9v battery to test and figure out which wire is which? I am going to wire it to some power supply that I have laying around. I don't have many electric tools, but I do have a multimeter I could test with but I don't know where to start. So extra parts that I have are: wires, 1 flipper switch, one power supply with the end still attached (haven't removed it yet), and one 2xAA battery pack.
Pic related is the bullshit I am dealing with. I don't remember why the straws are on those wires.

>>1334518
1.5V might not light a super bright blue led, voltage too low.
use a 1k to 10k resistor in series with your 9V battery and test with that so you don't fry your leds.

>>1334522
I do have a bunch of resistors, I forgot to mention that. I just checked the power supply and it looks like I did already strip the end off. The power supply is rated for 12V and 3A.

>>1334524
This is the package that the LED came in. I believe there are 6 of the LEDs already wired because there are 6 pieces of geode.

idots question about a capacitor:

what exactly does it do? just suck up extra V coming through the line and hold on to it? why not just put a resistor in if the current is too strong?

I want to get a switch to use as a dimmer. It would have a off position that has a click, and the on section would have a potentiometer. Is there a specific name for this type of thing/does it exist, or should I figure out how to do this with just a potentiometer?

>>1334565
Essentially, capacitor stores energy as a voltage. If you have a voltage across a capacitor, current flows through it and charges it up. when the voltage on the capacitor reaches the voltage it was hooked up to, the current stops. If the voltage the capacitor is connect to drops, like if it's short circuited, a current flows and the capacitor discharges.
the capacitor resists changes in voltage by making a current. If the voltage it's hooked up to doesn't change, there's no current. If you connect a discharged capacitor at 0 volts to a 12 volt source, the voltage rises from 0 to 12 very quickly. There's a spike of current, you'd get a big spark, and then the capacitor would be charged. 12 volts on the capacitor, 12 volts on the battery, and no more current flows because the capacitor is charged up to the same voltage.
So we can say that *the current through a capacitor is proportional to the rate of change of the voltage*.
AC voltage changes over time. The rate of change of an AC voltage depends on its frequency and its voltage. A higher voltage at the same frequency changes at a faster rate because it has to change more in the same amount of time.
A capacitor connected to an AC source at a frequency will draw a certain current for a certain voltage, just like a resistor. So much like a resistor, in fact, that it has ohms just like a resistor, except instead of resistance, it's called impedance. And unlike resistance, it depends on the frequency, because a higher frequency AC at the same voltage has a higher rate of change because the voltage has to change the same amount in less time.
So we can say that the impedance of a capacitor is proportional to the frequency.
This is how capacitors are used in filters. A capacitor connected between a signal source and ground has a low impedance at high frequencies and basically shorts out high frequencies, but has a high impedance to low frequencies and lets them pass. A high-pass filter.
That's capacitors.

>>1333747
could anyone be so kind as to help me figure out what the fuck is wrong with my pic code? i've tried to strip it down to the bare minimum. i'm just trying to toggle the ports on and off so i know the thing works. i'm using a chinese pickit2 and pk2cmd appears to think it's programming correctly. voltages are fine but i get no output on the pins.

https://pastebin.com/raw/W6EjzDCb

>>1334565
Essentially, capacitor stores energy as a voltage. If you have a voltage across a capacitor, current flows through it and charges it up. when the voltage on the capacitor reaches the voltage it was hooked up to, the current stops. If the voltage the capacitor is connect to drops, like if it's short circuited, a current flows and the capacitor discharges.
the capacitor resists changes in voltage by making a current. If the voltage it's hooked up to doesn't change, there's no current. If you connect a discharged capacitor at 0 volts to a 12 volt source, the voltage rises from 0 to 12 very quickly. There's a spike of current, you'd get a big spark, and then the capacitor would be charged. 12 volts on the capacitor, 12 volts on the battery, and no more current flows because the capacitor is charged up to the same voltage.
So we can say that *the current through a capacitor is proportional to the rate of change of the voltage*.
AC voltage changes over time. The rate of change of an AC voltage depends on its frequency and its voltage. A higher voltage at the same frequency changes at a faster rate because it has to change more in the same amount of time.
A capacitor connected to an AC source at a frequency will draw a certain current for a certain voltage, just like a resistor. So much like a resistor, in fact, that it has ohms just like a resistor, except instead of resistance, it's called impedance. And unlike resistance, it depends on the frequency, because a higher frequency AC at the same voltage has a higher rate of change because the voltage has to change the same amount in less time.
So we can say that the impedance of a capacitor is proportional to the frequency.
This is how capacitors are used in filters. A capacitor connected between a signal source and ground has a low impedance at high frequencies and basically shorts out high frequencies, but has a high impedance to low frequencies and lets them pass. A low-pass filter.
That's capacitors.

>>1334583
actually nevermind i think i've figured out that it is not programming correctly. i assumed it just always spat out "program memory errors" and that because "good" was populated rather than "bad" i was fine. evidently it doesn't actually do that when programmed correctly. guess i have to buy a fucking pickit3. this still isn't as obnoxious as altera.

Checked the open circuit voltage of this 12v 7ah AGM last night only to find it sitting at 11.8v, put it on the ctek xs 0.8 thinking I had forgotten to recharge and it came up good whizzing through the steps. After the charge indicator went green I took it off and remeasured, I was greeted with the open circuit voltage quickly collapsing to 12v and below where it finally settled around 11.8v again.

Do I have a shorted cell? The battery is approaching its 7th birthday and hasn't been abused (project battery), is this just normal ageing or has something else happened? The capacity has been getting noticeably lower and lower over the past 2 years.

I removed the top cover and two rubber caps shot off with a pop sound so pressure had been building inside the cell compartments, since its an AGM there is no liquid. With a 60mA load the voltage immediately drops to 10.9v if that gives any clues.

I've looked on youtube and its full of people filling up the cell compartments with distilled water and epsom salt, but there never seems to be a follow up video or capacity test. Is there any real chance of restoring it or is it a waste of time?

>>1334586
thanks for explaining that man, it made more sense here than trying to read other descriptions of it. i do a lot of hobby cap replacements on things and nice to know what they actually do.

Okay so I am trying to generate some high voltage just for the sake of it. I plan on making an H bridge using a couple of MOSFETs, then switching it alternatingly with 9VDC at a few kilohertz using a microcontroller (or a 555). Then I'll feed that shit into a transformers primary and boost it up to whatever the fuck I want
Am I missing something here? Thing should work right? I should be able to see pretty arcs right?

>>1334716
>9VDC
Make sure you use logic level fets then, most need 10v to turn on fully.

I've done it lots of time and it works fine.

I am a CS student at Univ. While my computer science education is alright, I feel severely handicapped not knowing much about electrical engineering. I hail from a family of engineers, my father, grandfather, uncle are all electrical engineers. I too, want to learn everything about electrical engineering. Every once in a while I see a random circuit diagram on the internet and get depressed because I cannot understand it.
Is there any resource through which I can study electrical engineering concepts, except an actual degree?

>>1334729
Youtube videos and forums, takes a while and you'll need to pester lots of people but its worth it. Try making some basic circuits as you go, even messing around with an LED, some resistors and multi meter helped me when I first started.

>>1334729
>>1334735
Afrotechmods, eevblog are two that spring to mind for real practical basics whilst electroboom can give you some crash course theory.

Stay away from kipkay, he's just a salesman and spreads misinformation.

So i want to get into repairing stuff and need a soldering station.
Is pic related sufficient as a first soldering station?
http://www.aoyue.eu/aoyue-int936-soldering-station-smd-soldering-iron-esd-safe.html
Or am i better of with the 50W version in the long run?
http://www.aoyue.eu/aoyue-int936a-soldering-station-smd-with-ptc-heating-element.html

So i want to get into repairing stuff and need a soldering station.

Is pic related sufficient as a first soldering station?
http://www.aoyue.eu/aoyue-int936-soldering-station-smd-soldering-iron-esd-safe.html

Or am i better of with the 50W version in the long run?
http://www.aoyue.eu/aoyue-int936a-soldering-station-smd-with-ptc-heating-element.html

>>1334612
>shorted cell?
Most likely. Give it to the recycler and get a new one. 12V 7Ah SLA (sealed lead acid) gel batteries are ubiquitous and cheap. AGM (absorbent glass mat) is not that common for this size.

>>1334753
>AGM (absorbent glass mat) is not that common for this size.
hmm, so what is this one then? I double checked inside the cell compartment with a light and non of them have liquid.

http://www.yuasabatteries.com/pdfs/NP_7_12_DataSheet.pdf

>"All NP batteries utilize Yuasa’s unique electrolyte suspension
system incorporating a microfine glass mat to retain the
maximum amount of electrolyte in the cells. The electrolyte is
retained in the separator material and there is no free
electrolyte to escape from the cells. No gels or other
contaminants are added."

What would suddenly cause it to short? Age? I guess its been subject to high current loads for brief periods during its life but never anything continuous.

>>1334751
I have this and it works great. Well, a similar one. Same brand.

>>1334751
This is the one I have:
https://www.amazon.com/Aoyue-Variable-Soldering-Station-Removable/dp/B00MCVCHJM
Not my first soldering iron, but my first actual station.

>>1334753
Sorry for all the questions but have you ever seen one of the caps pop off when the cover was removed? Is it normal for pressure to build up inside? Made one hell of a "pop".

>Give it to the recycler and get a new one.
Do they typically offer trade in deals?

Should add to this>>1334751 that the Int936 has 35W.
So the costs are 37,73€ for the Int936 w/ 35W
and 97,64€ for the Int936A w/ 50W.
I don't know if 35W will be enough though.

>>1334761
>>1334764
nice, unfortunately it's not available in Germany

>>1334769
sorry m8 didn't know you were in Germany
pretty much any soldering iron station will work though. most starter irons dont have adjustable temperature, so just try to get one that does.

>>1334769
https://www.banggood.com/YIHUA-908D-220V-LED-Digital-Display-Soldering-Station-Soldering-Iron-Kit-p-1059873.html
https://www.banggood.com/110V-220V-Inverter-Frequency-Change-Electric-936-Power-Soldering-Station-Iron-p-1093556.html
These are $25 and $40, shipping included. They take commonly available 900M series tips.

>>1334766
>caps pop off
Never experienced that, the four SLAs in the office UPS cannot be opened.
>trade in deals?
No, just because pic.

(I want my legacy captcha back)

I bought a TIG AC/DC welder and after starting to use it, it very quickly burned out the unshielded LED floodlights I was using in the garage. Is there anything reasonably easy to do to prevent this from happening?
I'd read that directly grounding all metals in the garage helps, but it's a lot of hassle and if that was the only option I'd just avoid using LEDs. Apparently some of these problems are caused by the high frequencies feeding back into the power grid; can I put a ferrite core choke on the LED's connecting wires (220V 50Hz AC) to mitigate that? And the last idea was putting some metal mesh around the LED to catch some of the RF that travels through the air from the arc, again not sure what frequencies I'm dealing with so this may not help much.

>>1334778
Well the caps were hidden under a piece of plastic that's glued down at the factory, its not really mean't to come of. I had to get a blade under it and pry up so the glue would break. Once that cover piece was off the caps were free to rise and it scared the shit out of me.

Do you work with these often? If so what's their typical lifespan in your application?

>>1334781
>its not really mean't to come of
it's not really meant to come off*

What're good job search terms? I'm a BS EE Minor Math, with a preference for embedded systems or programming. Technically I have some slight experience with research medical devices. I don't want shit to do with power systems or other civil stuff, and of course I want to work for video games.

I've tried debugger, programmer, not sure about medical systems, maybe 'medical engineer' might pull up something? I am a Burger living in McDonalds of course, New York. Spent almost a year programming for some telecom software (their fax side and build-system tools, C and Perl)

>>1334729
There's infinite books. Basic analog stuff is a good start. I was started with op-amps and kirchoff voltage and current laws. Op-amps are actually a good start because they introduce to basic concepts: assumptions, virtual grounds/relative values, moving on to real problems with the assumptions. Many basic books go over LEDs, blinkers, radio (which won't make much sense) and get into logic circuits which although useful, aren't that necessary until you seek digital things.

Microelectronic Circuits by Sedra Smith is figuratively and literally heavy if you get the book, but there's pdfs. I don't think it starts with op-amps tho.

>>1334781
>typical lifespan in your application?
Only have them in the UPS which does a periodic self test. Still full capacity after 3 years, mostly stand-by use of course. Don't know about the lifetime of the previous set because the UPC was bought used.

>>1334575
>switch with potentiometer
try potentiometer with switch

>>1334588
>no verify option
disgusting

>>1334729
Len Anderson K6LHA's "Radio Hobbyist's Designbook" is a good walkthrough, with an emphasis on radio. this is not a bad thing, necessarily, as plenty of digital these days runs at RF anyway, digital is really just a special case of analog, and radio is approximately everything you're not learning in computer architecture and logic design classes

>>1334766
>pop
probably hydrogen. it's normal for some pressure to build up. if too much builds up those caps are supposed to vent it. try (carefully) adding a bit of distilled H2O but don't expect much. probably time to replace it desu

>>1334780
just try stuff dude, you might learn something

I blew a 600mA fuse on my multimeter today. I was measuring current output of a USB microcontroller programmer. USB shouldn't put out more than 500mA, but I guess current from capacitances and USB before the current limit kicked in blew the fuse. Let that be a lesson to you guys.

I'm planning on having a PTC resettable fuse with a trip current of 500mA in series with a standard fuse with 600mA trip just in case. Anyone got any advice? My DMM has another fuse for 10A on a circuit that can measure down to 10s of mA, so losing accuracy in the low current circuit above 500mA isn't a big deal to me.

pro tip: always compare prices on ICs on aliexpress vs your local distributor. occasionally you will find the local price is cheaper than chink.
ex: ATTINY10-TSHR microcontroller
Digikey price: $0.35 / 1
Ali: couldn't find 'em for less than $0.45 each in 10s
ex: Si5351A-B clock generator
Digikey price: $0.92 / 1
Ali: best I can do is like $1.45 each in 5s

>>1335044
>let that be a lesson
yeah, the lesson is, always leave headroom when measuring and don't assume that "optional" amenities like current limit will be implemented or implemented well :^)
>PTC fuse
>got any advice?
yeah, don't. the linearity of your low-current measurements will go to shit. E=IR. surely you didn't think that PTC fuses "switch" between their operating points so sharply?

>>1335044
>measuring current output
By shorting the source? That's an unusual procedure.
>Let that be a lesson
Been there.. Wishful thinking will always get ya.

My Metex pocket DMM only has a single 400mA DC current range with a F800mA fuse, fortunately standard 5x20. It also has a PTC but I managed to blow the fuse nonetheless. I rarely use the mA range any more because I now have a set of low-Ω 1% resistors (1Ω, 100mΩ, 10mΩ) and measure the voltage drop instead. The result is more accurate because the series resistance of my meter (fixed test leads, ageing switch, PTC, fuse) is poorly defined which plays a role when you do circuits that run on a single cell and down to 1V. You always know the 'burden' voltage (silly term) because that's what you measure. Sense resistors are also useful for measuring (pulse) currents with a scope.

How is the average voltage/current of a waveform useful? I feel I understand why one would want to know RMS, but I'm not really sure what the significance is of the average.

So I actually built this thing. It is called a "Zapper" and it is basically a positively-offset square wave generator.

>>1335132

>>1335130
Circuits with sufficient filtering action react to the average value. This includes motors and solenoids. Similarly, LED and some other lamps produce light in relation to the average value of the input current.
Some components are dimensioned based on the average value.
Ye olde moving coil meters displayed average values, so it was important to know the average value of the measured waveform in relation to its amplitude, RMS value, etc.

Why are all online shops so fucking shit.
>looking for motors
>not a single website with a decent search function where I can just put my required values
>have to manually sift through hundreds of motors
>absolutely zero consistency in the way specs are presented, often even on the same site
>ozin, gcm, kgcm, Ncm, Nm, mNm
>rpm, rad/s, s/deg

F U C K
Y O U

>>1335159
https://www.maxonmotor.com/maxon/view/catalog/

I'd assume you'd rather keep using the shitty websites than pay Maxon's prices, though.

How would I do any of these? I've been trying to figure it out for days and I'm still not getting any of the answers right.

>>1335169
I am very bad at math so I can't help you

>>1335132
What is it supposed to do?

I need a 12 v dc power supply with a +,-, and ground, could I hook up one of these to a connector? It seems kinda dangerous though.
https://www.ebay.com/itm/New-DC-12V-8A-96W-Universal-Regulated-Switching-Power-Supply-Adapter-AC-110-220V/182987899957?hash=item2a9aedc435:g:WH8AAOSwL81aRgcz

>>1335189
Supposedly it is able to clean blood of bacteria, virii, and shit, but I was using it as the driver circuit on a 3-channel scalar interferometer that I built to combat the NWO. Unfortunately, I broke it and haven't made a replacement. I still have the Zapper though, just not the radionics device.

>>1335196
>12 v dc power supply with a +,-, and ground
Do you mean you need +12v, -12v, and 0v? Or just that you need 12v and a connection to earth ground?
One of these is not a dual supply, it's a single supply isolated from ground.
If the single supply is what you need then yeah you just hook your connector up to it. Dangerous? It even has a little cover for the screws. If it still bothers you, put it ina box.

>>1335196
I think you might need something more like this, if the -12V has enough current for you
https://www.jameco.com/z/RT-65B-MEAN-WELL-AC-to-DC-Power-Supply-Triple-Output-5-Volt-12-Volt-12-Volt-8-Amp-3-5-Amp-1-Amp-64-6-Watt_323540.html

>>1335215
Only the 5v rail is rated at 8 amps, the one he posted from ebay is rated 8 amps at 12 volts.

>>1335209
+12v, -12v, and 0v is what I need.
I wasn't sure where to look and thanks to >>1335215 I found this on jameco. It fits my needs perfectly and is in a nice package. Thanks for the help.

https://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?langId=-1&storeId=10001&productId=190960&catalogId=10001&CID=CATJAN218PDF

>>1335227
just to be safe, what are you trying to power?

>>1335233
Will do, it's just a drum machine. The power supply is kinda rare and it has this real wonky connector so I'm going to replace that.

>>1335242
that's unique. Also, the writing on that connector says 10 volts. you're sure it'll tolerate 12 volts?

>>1335250
I've read around and it's able to handle it.

>>1335252
should be fine, if the extra 2V aren't going to break anything

>>1335250
oh geez, it goes straight to two regulators and gets dropped to ±5V
it should work down to ±8V input and run cooler at that

>>1335255
pic belated

>>1335257
What about two diodes in series with your 12v feeds? Should drop it close enough to 10v.

>>1335262
could, but probably not much win in doing so imo

>>1334811
Ok, thanks for the help anon.

>>1334817
>try (carefully) adding a bit of distilled H2O but don't expect much.
I guess its worth a try and we do have a bottle of that kicking around somewhere.

Is that just enough to wet the plates but not too much that it submerges them? Will the battery be safe after that? It lives in my bedroom so don't want anymore pops or weird stuff happening.

Look at this shit I just opened, this is why you don't buy cheap power adapters on ebay.

>>1335308
absolute disgrace. this is why if you do buy cheap power adapters on ebay, you take them apart first.
I got a 48v PoE adapter and the shit wasn't soldered in, half the solder joints had just a faint hint of solder and the others weren't soldered at all. probably something went wrong with the wave soldering and they didn't notice or shipped it anyway.
worked fine after I soldered it though

>>1335308
Is that a paper clip?

>>1335320
Feels more like a clipped component lead but who knows.

>>1335313
The bottom feeder Chinese don't give a fuck if it works or kills someone, a big part of their culture now is fucking everyone over for a penny.

I've been given a whole bag of these by my grandma after a sort out, some of it looks ok but these two are definitely being parted out.

This one looks even worse, can you spot the isolation gap? I can't.

repostan' from the QTDTOT thread:

I'm looking for a way to keep the water tank in my van from freezing. I don't have a need for hot water on demand, but the tank freezing is a real pain. I was looking at link related: a 44 watt birdbath de-icer.

https://www.walmart.com/ip/Farm-Innovators-44-watt-Econo-Bird-Bath-Heater-De-icer/17718101

I have a house battery so all my electrical is 12v but I'm mostly trying to avoid using an inverter just for the heater. Can I wire the heater direct to the battery and run it off 12v DC or will that not work?

Or if someone has a more practical solution, googling for "van water heaters" and the like only returns full size hot water heaters.

>>1335358
Power in watts equals voltage squared divided by resistance. it's meant for 120 volts, so it has a resistance of about 327 ohms.
at 12 volts, that heating element will make 0.44 watts. not a lot. probably not enough.
you would be better off buying some kanthal or nichrome wire and winding your own element, that way you can make it exactly the wattage you need. resistance is just proportional to the length.

Hey guys I am going to ask a really stupid question, I am a second year EE student but I still don't know what I want to do with this major. I'll be honest and just admit that I picked this degree because "I like math" and didn't know what else to do. I like digital design at the moment but don't want to pursue something that I could have ended up in without an EE degree. What are the more math intensive fields of EE? Communications? I have a general ham radio license so that is something I could see myself doing I guess. Sorry for the retarded question.

>>1335358
>the water tank in my van
Is this an internal water tank you can easily modify, or something else? Because you can get a cheap 200 watt 12 volt submersible heating element to avoid the inverter, or build your own like >>1335366 says.

>>1335278
the electrolyte's supposed to be absorbed, right? then yeah just enough to touch the plates and provide some ions.
>in my bedroom
yeah nah you should get a new one. next time keep it topped up (charged) on a monthly basis while stored

>>1335308
I thought I was doing poorly when I got one with a solder blob gone loose inside the case, but that takes the kek
so solly you got the trainee's practice board

>>1335336
>a big part of their culture now is fucking everyone over for a penny.
they learned it from the best. remember their export-oriented industrialization started around the age of engineered obsolescence
otoh "sell on price, live on rice" and who the hell wants that

>>1335336
it's okay, the clear wire is at ground so don't need one

>>1335358
https://www.aliexpress.com/item/2M-12V-Self-Regulating-Heat-Trace-Cable-for-Freeze-Protection-on-Cars-and-Trucks-12V-Battery/32593349695.html
there are longer lengths in case you have exposed pipe that also needs protection

>>1335376
quit school, get your commercial radio license, see the world

>>1335336
>can you spot the isolation gap?
Yes, white marking.

>>1335336
right here

>>1335489
>>1335491
>>1335493
Here's the topside, they didn't use a safety capacitor across the transformer but an orange ceramic disk type.

The isolation is less than 1mm at its closest point.

>>1335489
I actually did keep it topped up and use a smart charger to maintain it about once a month. The others are still ok (albeit getting a bit old).

>>1335497
the only two discs I see are the primary snubber cap and the feedback coupling cap
damn that thing gives me the heebiejeebies. I wonder how many of these really bad commodity chinkshit devices are being assembled and/or designed as piecework by the otherwise unemployed, or kids snacking on sweets after homework

>>1335499
then it could be just plain old. don't expect much

>>1335515
You're right, it doesn't have a y capacitor installed. Hard to concentrate right now with all the building works going on.

>>1335515
Just noticed the battery actually has some bulges on the sides with extra stress marks on the casing. Will be monitoring the others from that era very closely from now on.

Anyone know where I can get the Apple IIgs keyboard schematic?

>>1335567
because it's a "y tho" capacitor

>>1335573
what do you expect to learn from it?
if you just want a usable keyboard, an old ADB Mac keyboard should work fine
if you're trying to play around with keyboard input, you would probably be better served finding the relevant ADB and keyboard specs and using a modern microcontroller

Why don't we have tons of conventions like the softwarefags get? They get things like Google I/O, devfest, mobile Congress, etc. The only hardware convention I can think of is makerfaire and it's more of a tradeshow than an event with speakers.

Why don't we have tons of conventions like the softwarefags get? They get things like Google I/O, devfest, mobile Congress, etc. The only hardware convention I can think of is makerfaire and it's more of a tradeshow than an event with speakers.

>>1335576
Thanks. The reason I was looking for the keyboard was due to one of the staff in our physics lab losing the keyboard to an Apple IIgs that they (still) used and my professor asked me if I could get a replacement.

I thought it would be a fun project to make one so looking at ADB keyboard schematics should help

>>1335616
They have them, gotta go to China, and all the people at the show want to jump off foxcomm buildings

>>1335616
>implying I/O etc. aren't trade fairs
where's the money gonna come from? the electronics hobbyist is not a profitable earner for anyone except maybe aoyue or vectorbord

>>1335635
detached keyboards are typically managed by a microcontroller that scans the matrix and turns switch closures into status reports to be sent over fewer wires than the raw keyboard matrix. Apple, post IIe, is no exception to this rule. you should definitely be thinking in terms of creating an ADB device because you're probably not going to get new Apple mask-programmed micros for love or money

>>1335308
If you need a cheap power adapter check your local thrift store. They always have shit tons of old adapters for like $2-5.

How difficult is it to make the move from using breadboard / stripboard to designing my own PCBs?
Considering that I have no experience with using any design program except using a freeview program to view gerbers.

Is there a simple program to start with, preferably that would let me just draw a line between components and then let me specify number of layers and do some sort of route finder type optimization?

>>1335754
you should try to get started with good habits.
1. not bothering with full autorouters, because they usually make bad and ugly decisions. most of them require a netlist anyway. pros generally don't bother with them because one generally ends up spending as much time cleaning up the computer's product as they would drawing one. they generally don't know enough about your circuit to make good decisions about keeping certain traces or components close or far apart
2. capturing your schematic first then working from a netlist generated from it. this makes circuits of even modest complexity much easier to work with. being able to see the rat's nest (white lines in Pic related) on the board helps you place and move components around, adjust board dimensions and shape, and makes the manual routing step so much more relaxing. you always know whether you are finished or not
I used to be an EAGLE fan but KiCAD just rocks the socks. I recommend it to anyone. it's got a few quirks and oddities, like the discrete step of component-to-footprint association which isn't always very smart or automatic, or the component/footprint librarians which are a bit rough, but otherwise it does the job well, it's free, its libraries are decent, and it doesn't have any annoying artificial size/component limitations (not sure if you can use more than 16 layers, never tried). you can set up a separate autorouter if you really want to. bonus: fapping vigorously to 3D rendering of your board

>mix up vdd and gnd on usb connector and fry my pic
i do shit like this all the time yet somehow never learn

>>1335754
the actual work behind designing a pcb is very simple if you're not doing switchers or high frequencies. learning to use clunky leviathans like altium for the first time will make you want to blow your brains out though. i put some of the blame for the above greentext on having to define a usb port schematic symbol and then create a pcb library file which references the numerical designators on that symbol rather than the pin names.

does anybody know what this material is? i would say nichrome but its all black and unsmooth.

trying to repair this samovar without having to buy another element.

>>1335869
Nichrome does turn dark like that, but it might be Kanthal, which is preferable for high-temperature uses.

>>1335879
ohhhh, both these are chrome alloys. internet says i need silver solder, but i think ~450F might be too low a melting point for a heating element. am i just being retarded?

>>1335886
I'd crimp their ends together with high-temperature trim connectors instead of soldering, that's probably how they're affixed at either end anyways. That or a ceramic terminal block.

>>1328818
And lo, he reached out his hand touching the thread and it withered and died.

>>1335767
Is there a straight-3d PCB design application? Seems like it'd be an interesting testcase for VR or just a "3d modeler" application, just with a whole lot of autocad-like behavior and then you could throw an algorithm at it and see if it makes good traces then export to an etcher/cutter/whatever

>>1335904
are you complaining about slowness on /diy/? the board where a crab meat thread lingered on it for like 6 months?

>>1335916
Well I use sketchup, but I don't think it's what you're looking for.

>>1335711
No need, my grandma dropped off a load of unwanted adapters so I'm just getting rid of the crap ones.

>>1335886
Stuff with 450F melting point would be some silver containing tin alloy, while "silver solder" would refer to some brazing alloy. They have much higher melting points.
I'd do what >>1335890 suggested, though.

>>1336064
why not just put little bends in each one and then hook them over each other?

>>1336083
They probably get too soft at high temperatures for that to work well.

>>1335916
no, anon, this is engineering, not >>>/v/
if you want to """"""innovate"""""", >>>/g/
if you want to route boards, study some layouts on the net and maybe actually route a few boards to get a frame of reference

Building my own rover bot. Pretty set on using a 3S LiPo, and a bunch of different motors and controllers which all take different voltages (RPi 5V, Arduino 6-12V, a couple of steppers tbd, and two 12V BDC motors). Also some sensors and cameras. The arduino will be reading sensors and driving motors, the raspberry pi will be doing machine vision and telling the arduino where to go.

What would be the best way to go about powering everything? Was thinking about connecting the motors (drivers) and pump directly to the battery's Vcc, maybe the steppers too if I find some that take ~12V and are still small enough (they don't need to be very powerful).
Then have a buck converter with 5V 2.5A output to power the pi and connect the arduino to the pi via USB. I would use the GPIO pins to send signals to the motor drivers or receive data from the sensors, and connect the sensors' Vin either to the arduino pins or the 5V line if they draw >40mA.
Lastly I would use the on-board wifi on the RPi to reprogram both it and the arduino, if needed.

Questions:
>Does all this make sense?
>Should I make my own power distribution board? I have no idea how to go from the battery's xt60 connector to something that can power the raspberry pi (except for soldering cables to each other like some sort of savage).
>Is it a good idea to have the motors and controllers on separate lines (electrical noise)? Should I solder caps on the motors or does it depend on the motor/driver and such?
>What about the ground, do I just connect all the grounds together, or have different ground pins for each voltage level?
>Also, can I just use the battery's charge balance cable to monitor its voltage and tell the robot to shut down when it gets too low?
>What about trying to disconnect the circuit in case of short circuit/overcurrent, is it even worthwhile? These batteries can easily output well over 100A

Is this a good plan? It's my first time building something like this so I'm not really sure.

>>1336153
>motors (drivers) and pump directly to the battery's Vcc
strictly speaking, positive supply, not Vcc, which construct only applies to bipolar circuits (cc = collector, likewise Vdd = drain, Vss = source, occasionally you'll see Vee = emitter) but yes, avoid power conversion where you can
>buck converter with 5V 2.5A output to power the pi
good
>and connect the arduino to the pi via USB
would work, in theory you could also inject 5V directly into the shield ports
>on-board wifi on the RPi
sure
>from the battery's xt60 connector
obviously, soldering/crimping wires to a mating connector.
>like some sort of savage
civilised makers use one or more of Pic related with jumpers on one side, loads on the other, supply on either side
>Is it a good idea to have the motors and controllers on separate lines
star topology for power supplies is good practice. definitely for ground but sensors may have enough of their own supply filtering. you can always jab a disc cap into the terminal strip if it proves helpful
>can I just use the battery's charge balance cable to monitor its voltage
you could, using the top output avoids resistive loss of the power cable. pick an analog input somewhere, and resistively divide Vbat to input range with a few hundred kOhms. the individual cells don't really matter in discharge. note, battery voltage isn't linearly proportional to state of charge, so when the voltage drops below 3.5V/cell, it's about to dive quickly and your bot should consider its response e.g. heading back to base. ideally you should draw 0 from the battery below ~3.2V to avoid wear/damage
>What about trying to disconnect the circuit in case of short circuit/overcurrent, is it even worthwhile?
absolutely. put a fuse between the battery's + terminal and your stuff. drone packs do have built-in protection but you may want more sensitivity
also, thanks for awesome new bread pic

Hey guys, I have a question regarding transformers:
I have a salvaged toroidal one with 2* (19V, 60 VA) secondaries, wich I plan on wiring in parallel to drive a low resistance heating element (1-2 Ohm) for a soldering iron. A microcontroller with back to back mosfets will manage whether or not the next half cycle of the mains will drive or not the heater.

Obviously, the current through the transformer will far exceed the rated value when the heater is on. though the microcontroller will be able to keep the mean maximum within the allowed range, I am not sure regulating at only 100Hz will be sufficient to avoid damaging the transformer in the long run.
I should probably do some thermodynamics and calculate the heating power during one phase with the heating element on, but wanted some input. Any idea about this?

>>1336167
Thanks for the pointers. Just to be clear, for the raspberry pi connection would something like pic related be a good solution?

>you can always jab a disc cap into the terminal strip if it proves helpful
I'm not sure I understand this part. You mean put a small cap in parallel with each sensor?

>>1336176
When you do your 'thermodynamics', remember that a good soldering iron has a rather thin and highly flexible silicone cord with a typical length of about 1.5 meter. Good luck.

>>1336176
what about PWM and an in-line inductor, buck regulator style? you might need to increase the PWM rate well over 100Hz for that to work but it would be fairly precise
time to heat up should be fairly short. monitor the temperature of the transformer itself with a sensitive thermometer, power transformers shouldn't get more than about 30°C over room temp

>>1336185
micro USB is a terrible high-current connector. if there's a barrel jack that takes 5V, use it instead. if not, look into feeding power to the pin headers (some SBCs will accept this, some will not)
also, you probably want star topology. see cap in Pic related, but it's probably better to install it at the sensor end of the wire than the supply end as drawn

>>1336205
>look into feeding power to the pin headers
I think I read somewhere that doing that is discouraged because then you bypass the safety circuit built into the pi. On the other hand it says in the documentation that the microUSB port is rated for 2.5A. Since that seems to be enough for me I thought I'd go with the microUSB. I'll do more research though.

Thanks for the clarification about the cap and nice drawing, the pie looks legit.

>>1336218
>bypass the safety circuit
that's a good point. I also understand the pi has built-in mitigation for power droop than most devices, so micro USB should be fine. use a short cable, preferably thick
>wire for each cell required?
just one, the same one that connects to the + output of the pack. in fact, you could use the + output of the pack if you tap it very close to the connector. there's nothing you can do about balancing while discharging, anyway
btw you could also feed that battery level input to the lolduino if you have enough spare analog inputs, not sure if the pi even has any

Something I thought I'd run by you guys. Company down the hall from me installed a magic box in the electrical room yesterday:
https://voltserver.com/idea/?gclid=Cj0KCQiAzrTUBRCnARIsAL0mqcyA2MVgpne28tzqgiEqo0PXgri0T6IRMpg3dLB5lia6nT9jTgWCrU0aAhgqEALw_wcB
This sounds like power over ethernet. Anyone see why these guys are trying to reinvent the wheel? Are they even trying to (presumably the end device would have to conform to a standard)?

>>1336222
>+ output of the pack
Yeah that would be my backup plan, but I've read that the JST connectors on LiPos is used so that the charger can read the voltage of each cell, which is pretty much want I want to do, just without the charging.
>not sure if the pi even has any
You're right, that's gonna have to go into the arduino. I think there are some ways to do analog on the pi but IIRC it's a bit hacky and in this case not really necessary so I won't bother.

For the life of me I can't find a socket for this IC to fit in to. I'm looking for a through-hole socket for this IC so I can solder the socket to a custom pcb and then into a breadboard;
https://media.digikey.com/pdf/Data%20Sheets/Texas%20Instruments%20PDFs/CC2530F32,64,128,256_Rev2011.pdf
I keep running into 44 pin sockets but no 40 pin sockets. Understandably, 40 pin ICs are probably uncommon.
Worst case scenario I solder the IC directly to the custom pcb, but I'd rather not go through that headache of eventually having to rip it off the board and transferring it to a developed pcb.

>>1336250
>The technology combines energy and data into energy packets and transfers hundreds of packets each second from a VoltServer transmitter unit to a receiver unit.
topkek, is RI a medical marijuana state
they're not exactly reinventing the wheel so much as streamlining it, apparently doing with one pair what PoE does with 3 or 4. it reminds me a little of how caller ID is done.
>standard
nah, it's patented. if there is a standard, they'll be the ones writing it

>>1336255
not only to read, but also to equalize by shunting charging current around cells that are fully charged. I just don't see what would be actionable about individual cell voltages, unless you were doing charge control within your setup. individual cell voltages won't give you more information on the state of charge than the total Vout would, in fact, discharging the cells unevenly through the voltage divider isn't going to help service life any

>>1336263
silly anon, that's a QFN. you don't socket those, much less if they're RF circuits
>TI provides a compact reference design that should be followed very closely.
you need to design a testing/development board use the debug interface to program the flash
when your design is complete, just buy a new chip, use the debug/programming interface to program the flash just like you did on your testing board, and put it on the finished design board. it's only a $5 chip

>>1336250
Covert Product Placement? Do they pay well?
https://voltserver.com/idea/ leads to the same page
gclid=Cj0KCQ.. is your personal session id
Referral linking is not welcome
4chan.org/rules#global

>>1336275
You'll have to forgive me for my naivete, never worked with QFNs.
For anyone that's interested, I found that Texas Instruments sells their own dev boards for the chip in question, but they look incredibly involved and expensive;
http://www.ti.com/lit/ug/swru208b/swru208b.pdf
I did stumble upon a breakout board though that meets the correct dimensions;
http://www.chipquik.com/datasheets/PA0071.pdf
Think I'm just gonna go with that.

>>1336304
Nah, dirty phone poster who couldn't be bothered to edit his copypasted URL.

Anyone know of any good Boolean algebra resources? I am trying to learn from the book Digital Design by M. Mano but I'm having a tough time.

>>1336333
>Anyone know of any good Boolean algebra resources?
ROPE
-t. actual brainlet who spent 3 hours on one problem last night

>>1336356

i know that feel anon

>>1336312
>never worked with QFNs
ohshitnigga.jpg
welcome to the New Electronics! you're gonna have fun trying to solder QFNs. you'll need a solder paste stencil, solder paste, and a hot air station, and a hot plate would really help. you will quickly lose your hair otherwise
if this were not an RF device, that ChipQuik breakout would be fine. BUT those RF traces need to be impedance-controlled and are sensitive to layout. the resulting stray reactances are on the same order as the components specified in the matching network
I'd suggest downloading the design files from the CC2530EM page but they need some significant massaging (add tool rack, rename files) before they're ready to submit to a board fab
so I'll suggest you use a specific breakout board with all the RF crap handled such as https://www.aliexpress.com/item/ZigBee-Wireless-Module-CC2530-Module-Internet-Of-Things-Core-Board/32503115283.html

>>1336333
K. Roth's "Fundamentals of Logic Design" worked well for me, I don't even know if it's still in print

>>1336333
I'm a Digital Fundamentals (Floyd) kinda guy, but what sort of stuff are you doing? It covers from the basics of gates and truth tables through algebra, Karnaugh maps, and some primitive PLAs and memory arrays.

>>1336356
>tfw never had to look at the textbook because I learnt boolean algebra from playing minecraft
>got an A+

I'm trying to make a flyback converter, but the output voltage is about x10 greater than I expected. It looks like I'm getting ringing (I think that's the term?) on the switch. What is graphed is the node where the switch and primary winding meet.
What do?

>>1336406
First of all, use a FET instead of a switch, assuming there's one with a high enough forward voltage. Secondly, where's your feedback? For a simple flyback converter you'll need a comparator running off a voltage divider on the output winding that triggers when the output voltage is high enough and pulls the switch down to ground. Of course you'll need to power it some way (another winding?) and you may want to use an optocoupler to keep it isolated. But since it's grounded on both sides anyways it doesn't really matter.

>>1336416
Got to start small, man. I'm just building the most basic design first, then I can easily check if my math is right for t_on, number of windings, etc.

>>1336406
if I read correctly your duty cycle is 75% or so and your transformer is a 1:6 step-up(?)(inductor math always fucked my shit up). not sure why you would expect much less than 80V output under those conditions
try adding a snubber to the primary side and maybe scale the inductors up in case they're "saturating"
>a 1N4148 instead of a Schottky for power rectification
shiggy

>>1336418
>if my math is right
what did your calculation predict, care to post it?
and is the load really 5Ω?

>>1336397
I'm currently working through M. Mano's Digital Design. However, the explanations in this book are HIGHLY abstract and it basically leaves the practical application of the provided theorems and postulates all up to the reader.

I should note that I am perusing this independent of an actual class and have no instructor to supplement the book.

>>1336418
If your math is just assuming that Vin/Vout = Nin/Nout (or the square/root of the inductance ratios), this doesn't work for flyback converters, they're essentially just a boost converter with the output being magnetically coupled instead of directly connected. The primary voltage spikes like your image and that's how boost converters increase their voltage. This multiplication factor is dependant on the resistances in your circuit, and this includes the load, so if you want a regulated PSU at all, feedback is necessary. If you turn that 1mΩ switch into a 1µΩ switch, chances are you'll see an even higher voltage on the primary and secondary. You need feedback, but your feedback can work by varying pulse length instead of a simple bang-bang converter. Needless to say, this is harder to accomplish with an isolated PSU.

>>1336430
That could definitely be an issue. I recommend buying a bunch of simple 7400 series logic gates and a breadboard and messing about with them.

>>1336397
I'm so retarded that I have 2000 hours in garry's mod making wiremod shit with gates, and I never bothered to learn it.

How do I read AC circuit diagrams? AC goes back and forth so how do I understand what's happening in a circuit? With DC it's just sequential but with circuits like pic related idk what's going on.

>>1336726
Input AC voltage appears at the gate and modulates the drain-source current. Changing current causes changing voltage drop across drain resistor(s) which appears as amplified signal at the output.

Pic is more like an AF amp, RF usually would have smaller capacitors. If you leave off the caps you have a DC amp.

>>1336742
Why are there two inputs though and why a resistor across them?

>>1336746
lowest line is ground (0V) not input

>>1336757
Nah on the left there's two inputs

>>1336778
two halves of the same input. voltage has to be relative to something just as you can't push without something to push _against_
the resistor is actually across the gate and source of the field-effect transistor, to keep the FET's p-n junction reverse-biased and therefore able to effect a field to modulate d-s current flow

>>1336778
And on the right there's two outputs

>>1336784
two halves of the same output.
hey let's really blow your mind right the fuck open and observe that the "one" +9V input at the top is only half of an input!
just as with the input, voltage doesn't mean much without something to compare against

NEW BREAD

>>1336788
>>1336788
>>1336788

NEW BREAD