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



Thread archived.
You cannot reply anymore.



File: 1509166220011.jpg (262 KB, 1280x960)
262 KB
262 KB JPG
>>1299048 antiqued thread

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

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

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

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

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

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

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

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

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

>My circuit doesn't work. Halp?
Check wiring, soldering, part pinouts, and board artwork if applicable, then post schematic. Supply ALL relevant info and component values.when asking a question.
>Li+ batteries
Handle with extreme care, or... https://www.youtube.com/watch?v=gz3hCqjk4yc
>>
is it possible to take a bunch of used speakers from various stereo systems and interlink them in a giant plaster ghoul wriath with LEDs? I'm mostly worried about the speakers being compatible and audio being broadcasted through a dozen or so speakers.
>>
File: 1505287892734.png (100 KB, 1212x553)
100 KB
100 KB PNG
this thread's digits brought to you in part by the LTC3811 dual phase high-power buck controller
>>
>>1303820
>possible
sure
>will they sound good
nah
>will they even match
probably not
>will the LEDs blow when any worthwhile amount of power is put through the speakers
probably
you're better off doing the speaker amplifier(s) and LED driver separately. the exact arrangement depends on the impedance mix of the speakers you happen to have
>>
>>1303822
>will they even match
never even thought about this. What can I do with old speakers?
>>
>>1303821
I like this tradition.
>>
>>1303809
>however you want to slice it, 200mA * 180V = 36W. boost converters are not very efficient at high boost ratios but 42% sounds bretty bad.

Well, do you have any other suggestions? I'll admit I've run into several design walls trying to create the aforementioned boost converter namely relating to the efficiency as you mentioned and high the primary current draw when the load is drawing maximum power. That's why I wanted to use a flyback driver which I've heard were good to provide 100-200W of power but in practice they seem to have similar limitations to regular boost converters.
>>
>>1303636
>its just the same thing as the fucking voltage meter

the voltage reading on a battery is meaningless unless the battery is under load. so the ''battery'' ranges put a load across the battery. you could've discovered this yourself by experimentation, by noting the readings are different, but that's such hard work.
>>
>>1303576
>>1303611
any ideas?
>>
>>1303860
buy new ic's
test diodes and other passives
check connections and trace tracks.

in other words, faultseek.
>>
>>1303881
OK i just wanted to make sure my intuition wasnt wrong and to check if the ICs had a lower/higher chance of failing vs other parts. my first thought was that an IC was fucked but i cant tell since nothing is visibly damaged and i was worried i'd be chasing ghosts by trying to replace the ICs when XYZ component might have a higher chance of failing than an IC. got all my parts coming in from china express already but its gonna be 20+ days before my supply can be fixed so i went ahead and ordered another $15 buck/boost converter from ebay that should get here in under 15 days so i can once again have a variable power supply.

if it makes a difference, this all happened because i was fucking around and shorting wires to make sparks and shit and now the onboard voltmeter never reads anything under 10V (the shield itself isnt damaged i dont think and the buttons work fine or so it seems since i can still 'adjust' the power settings but the voltmeter itself never will read under 10V). im pretty sure its a voltmeter and not just some cheapo setting cause my multimeter will read what the onboard voltmeter says but adjusting the numbers (the numbers will change but then it will show the voltmeter reading after adjusting the setting) has no effect under 10V. it works above 10V but the setting is also like permanently offset from the real output by like 2V as well (if I set it at 12V, the output will be 14V, if its set at <10V, the output is always 10V).

until i get the new stuff though it looks like im gonna be using a pot to get under 10v and constantly testing against my multimeter to get proper voltages.
>>
>>1303889
not him but there's at least three different configurations for buck boost converters that come to mind that will all have different failure characteristics. it looks vaguely like your board is just a boost converter followed by a buck converter. if that's the case, you should find the buck fet and use your meter to probe from its gate to source measuring frequency/duty and then dc voltage if there's no meaningful waveform there. that fet blocks the output when off so either it's failed short or some parts of the control circuit still work. also feel how hot the heatsinks are at no load.
>>
>>1303969
sorry i'm retarded and didn't read half your post about the meter being the only broken part
>>
File: 1494659539081.jpg (18 KB, 558x335)
18 KB
18 KB JPG
>>1303833
yes. it's important to know what's out there.

>>1303844
low-resistance FETs, probably a much larger inductor. perhaps try a lower frequency instead of a higher one, so the inductor builds more "momentum". pic related
>>
>>1303969
interesting. i'd try that out but i dont have an oscope so i dont think i can really look at waveforms on the mosfet. but thanks for the insight, i dont know shit about boost/buck converters which is obviously why im buying them off ebay but hopefully ill know enough soon to be able to construct my own or at least fix the one i have now.

>>1303970
assuming youre the guy i just responded to, i dont think the meter is the only broken part. it constantly outputs 10V even when its set to OFF so something else besides the voltmeter is broken i think. i dont even know if its a true voltmeter, the shield for this board is the voltmeter im pretty sure and it seems to function normally (it retains memory of the current voltage setting which i can change but it doesnt actually change the output voltage). the voltage setting works above 10V but its also offset by 2V at that point so it still doesnt work 100%. if the voltage is set under 10V, it wont output anything except 10V.
>>
File: up-down-conv.jpg (48 KB, 624x482)
48 KB
48 KB JPG
>>1303433
As delivered this thing does 2-24V to 12V but you can modify the divider to get 9V. Currently values are 9312 and 1002 which points to a ref voltage of 1.164V. If you parallel 2433 to 9312 you will get about 9V. Efficiency will be abysmal and the battery will not last that much longer, if at all.

What is still missing is a bost converter that sleeps until it needs to boost. 9V devices are often milliamp customers that could be supplied by a tiny regulated charge pump converter, something to clip between battery and clip, in 5V-9V out 9V. Do you listen, Chinaman.
>>
>>1303976
your multimeter probably measures frequency and duty (% on time) if it's not really cheap. probe with those settings between the left and right pins of each FET and see if you get anything. if you don't have those just probe DC voltage.
>>
File: 1492973453004.jpg (118 KB, 640x640)
118 KB
118 KB JPG
>>1303995
Chinaman sort of hear you! Ask louder!
>>
>>1304001

what are those? kinda look like voltage regulators but kinda don't.
>>
>>1304028

well fuck. nobody's going to tell me now that I forgot and left my joke name on from the physics symposium.
>>
File: lm.png (67 KB, 722x269)
67 KB
67 KB PNG
>>1303995
Thanks, I've been thinking of going with LM27313. I think it looks kind of promising. I might get one or two of those from TI as they have free shipping this week.

But yeah, I do not know how this chip is going to work when the 9 V battery is new and its voltage is equal or above the regulator output voltage.
>>
Hello
I have a piece of equipment from 1977 that im trying to repair. I suspect a diode, but im not sure. Does the forward voltage drop change its value with age?
>>
>>1304038
As a general rule, boost converts can't regulate their output if input is above the nominal output voltage. So, if your battery is super fresh, the output will be slightly above 9V.
If that's a problem, raise the output voltage a bit or use different topology, like SEPIC.
>>
>>1304043
No, and diodes normally fail either by short or open circuit.
Are you measuring it properly?
>>
>>1304044
lower as in Vin minus Vd, isn't it, unless load current is bursty?
>>
>>1304052
Yeah.
>>
>>1303996
do i need to remove the mosfet from the circuit before doing this?
>>
>>1304081
no, you should do it while it's powered on
>>
>>1304038
>free shipping this week.
again? geez
it would be nice if digi-key did this for all their orders one week and not just for their sample stores
>>
>>1304121
or you could just order from aliexpress
>>
>>1303996
>>1304082
ok so what should i be looking for? i think my meter has a freq option (labeled hFE). you said to look for a 'meaningful waveform' so im just looking for something other than just nil?

also i dont remember if you were the same person to test the diodes but do i need to remove the diodes from the circuit to test them? I tried testing the SMCs and the tiny little diodes but i wasnt getting a reading on anything, not even the SMC resistors
>>
>>1304214
hFE is a meme setting that they put in as an inside joke. your meter doesn't have a frequency range apparently. in that case you'll just want to measure the DC and AC voltage between the two outer FET pins to determine if it's stuck on or switching at all. an AC reading above zeroish means the FET is switching. a DC reading above ~3V with no AC reading means it's stuck on.

all meters have a "forward voltage' range with a diode symbol. use that, and measure both directions. the diode doesn't need to be removed since we just care if it's short or open. if it's working you should get 0.3-0.7V one direction and nothing the other. i didn't suggest you do that though so maybe another anon had something else in mind. i don't think any diodes in the power path fried though because you'd be getting an AC output rather than a steady 10V. in fact, measure the output on your AC voltage range and see what it reads.
>>
>>1304221
yeah i dont think anything else on my meter would be a frequency meter. its a $25 meter from wal mart. i suppose my next thing to upgrade is my multimeter but this new power supply is eating into my hobby money so im gonna have to put that off for a minute.

thanks ill test that later ive got some other things i need to take care of first but thanks for the help.
>>
>>1304214
hFE isn't frequency, it's measuring DC gain of a bipolar transistor.
>>
>>1304028
>>1304030

They're cheapo 3.3V regulators with charge pump to double the input voltage so they can work with anything from 1.8 to 5V in. Uses a chip similar to http://cds.linear.com/docs/en/datasheet/3240fb.pdf
>>
I've got a boost converter that needs to go to a fairly high voltage, and it isn't reaching that. As far as I know, decreasing inductance and frequency will both increase the maximum voltage, anything else I should know?
>>
>>1304327
i just had this problem. increase your frequency before touching the inductor, it's possible it's just starting to saturate because your on-time is too long. also post schematic.
>>
>>1304334
Increase the frequency? Won't that increase the losses via transistor? If it is reaching saturation then I can see why I'd have to increase it, but I set it at 24kHz to decrease ripple as much as possible so I don't think that's the issue. Frequency probably is too high. I'll work on it tomorrow and see if I can get it running, but I removed it from my breadboard last time since I needed it for other prototyping.

Thanks for the response, I saw some boost converter stuff going on last thread so it got me working on the project again. I'm doing feedback through a voltage divider feeding a comparator off the high voltage rail, not a zener diode/other method which I think some people do. Using an IRF840 FET.
>>
>>1304340

>Won't that increase the losses via transistor?

Yes.

>but I set it at 24kHz to decrease ripple as much as possible

Higher frequency will generally reduce ripple. Sure you calculated your inductor correctly for 24kHz?

24kHz is pretty low.

>my breadboard

Keep in mind the parasitic capacitance and inductance, they can be pretty significant.
>>
>>1304340
24kHz is on the lower end of frequencies and i'd bet you anything your resistive losses swamp your switching and coss losses unless you're driving your fet with a resistor to pull it up/down.
>>
>>1304343
>calculated your inductor
Nothing of the sort happened. I used the biggest one I could find (1mH), but now I don't think that was a good idea. At all.
>>
>>1304359
a 1mH inductor as 24kHz is reasonable if it's a big toroid or equivalent, not if it's an axial. it's going to be much easier to advise further though with a picture or schematic.
>>
>>1304365
Axial. What is the effective difference?
>>
>>1304381
there's no inherent difference but axials tend to be smaller and have finer windings, which means both a higher resistance and a lower saturation current. from what i can see nothing else in your circuit could be expected to cause the problem you're getting. you could also modify your 555 circuit to give an ~80% duty cycle (at a higher frequency, more than doubled) to increase the output current capability assuming your inductor can handle that. your gate driver is good. i'd decrease the pnp base pulldown to 1k if you increase the frequency.

why do you nerds insist on bang bang converters? control loops are fun.
>>
anyone got any cool little things theyve made recently? need some inspiration for a new project.
>>
File: sdfdfs.jpg (815 KB, 1071x1000)
815 KB
815 KB JPG
>>1304393
i got this in recently but i haven't assembled it yet because my paste went bad. it's a FET bridge for a high current low voltage mains transformer. the benefits are decreased power dissipation and voltage drop over a standard bridge. it also includes a switch between center tapped / full bridge configurations for voltage doubling. it's not really useful to me yet but i liked the idea of it.
>>
>>1304385
>why do you nerds insist on bang bang converters?
It's my first boost converter, give a guy some space. Do axial chokes tend to have worse hysteresis too?
>>
>>1304399
honestly i don't even know if hysteresis matters in a boost converter because you're only operating in one quadrant. anyway your inductor in your circuit will have a nominal peak current of 250mA, which isn't a lot to ask of your inductor (29uJ nominal per cycle). so unless your inductor is particularly small (less than 0.5cm3? that's the smallest rated inductor on digi) then i'm no longer sure what your problem is. does your converter fail to reach its target voltage even with nothing but your dummy load?
>>
>>1304200
I regularly do. if I only wanted what aliexpress had that would work out nicely

>>1304381
I don't like that BJT as a transmission gate. your enable signal may be turning on the FET even when the 55 output is low. instead, wire that feedback into the 555's controls somehow, either pulling RST low when the output's too hot or replacing that comparator with an op amp to create an error voltage feeding back to the CV.
>>
>>1304398
Why not just http://www.ti.com/lit/ds/symlink/sm74611.pdf ?
>>
>>1304423
insufficient reverse voltage, and is actually more expensive than each ideal diode circuit block on that board. it also doesn't seem to have any data on its switching speed so since it's intended for photovoltaics where it won't be toggling at 60Hz i wouldn't feel safe using it. might work fine though.
>>
>>1304426
It's a neat component in any case, but no clue how it works.
>>
File: file.png (27 KB, 314x271)
27 KB
27 KB PNG
>>1304433
it is. some guy posted it here a while ago thinking it was a magic diode. it's really an integrated circuit combined with a power mosfet. i'm not sure how their charge pump works but mine just stores energy from the reverse biased half cycle of the AC input. they don't have that luxury with solar cells.
>>
>>1304438
It works in bursts. That is, it alternates between an ideal diode and a normal diode.
>>
>>1304322
>They're cheapo 3.3V regulators with charge pump to double the input voltage so they can work with anything from 1.8 to 5V in.

Thanks for the reply. I need one of those on occasion.
>>
>>1304381
>that circuit
Is this part of the 12V -> 180V/0.1A project?
>>
Is there any kind of ruler in Kicad's footprint viewer?
>>
does anyone know what kind of cable this is? the core is either aluminum or tinned copper, and it looks like it's shielded. on one end of the cable, the shielding has been soldered to the green/white wire of a cat5e cable.
>>
Hello /ohm/, I come to you with a noob question.

I was asked by my father to get a couple of car batteries because him and his buddies are trying to power up an old warehouse to use as their little retirement club and discuss about wars, wives, illegal immigrants and hookers.

Now, I know shit about electronics, and my father knows less than me, the only thing I know so far is that they plan on powering up a few light bulbs and maybe a couple of small appliances with an inverter rated for 3kW (completely overkill, but they get it for free so they want to use it), turning 12V into 220V.

Now, he has no idea what kind of batteries he needs and neither do I, but I can get any battery I want because I used to be a car mechanic and my old boss lets me use his discounts when I buy car parts.
Do I just go for the "the bigger the better" logic, or is there some kind of specific value (either Ah or CCA) I have to figure out to get the best performance?

They plan on buying two batteries and use them one at a time, while the other sits at home hooked up to a trickle charger to swap when needed.

Thanks a lot
>>
>>1304536

That looks like a cable you would see with higher voltage. Low voltage stuff is rarely soldered, unless you're a complete noob. On a cat5 makes me think it's used for PoE but I'd have to know more about the setup. Could just be some janky twisted pair landline setup.
>>
>>1304554
it's in a truck, it's used for a satellite phone. I think it's used to shield the cat5 cable, I don't think the cat5 cable has any shielding from itself.
it's about 1mm^2 thick, so I doubt it's for high voltage.
>>
>>1304542
use deep cycle batteries, NOT car starting batteries. like marine, golf cart, RV.
as big as you can get em or as big as they want to lift. more Ah means more run time.
you don't want high CCA. cranking amps are for starting cars and are not meant to be discharged, discharging them is bad.
>>
>>1304438
The data sheet says how it works, it charges from the bypass diode voltage drop and it toggles between fet/diode to recharge the stored energy capacitor as needed. 99.5% duty cycle 8A @ 25C
Pretty clever little device.
>>
>>1304567
Those batteries would be way more difficult (and probably expensive) to find compared to relatively inexpensive car batteries.

As far as I know, they need to power a few LED bulbs, charge a bunch of phones and maybe hook up a laptop charger once in a while, they will probably never even draw half a kW in the busiest days. And this will happen once or twice a week, disconnecting everything when they leave. It's a very low cost, low effort solution.

The warehouse is currently hooked to grid power and they don't want to pay the ~50€ a month it costs them just to have the service (so it's more, including the cost of electricity itself) when they use the place for such a short time, so things need to be cheap or they can simply bite the bullet and stay on grid.

Are car batteries really that bad? is there any other kind of battery that may be better for this purpose, like gel car batteries? If a battery breaks in the span of a few months then it's an acceptable trade for them, considering I can get a decently sized battery (around 110Ah, typical for a medium sized diesel engine) for almost as much as they pay for one month of grid service.

I just want to be sure that things don't go kaboom or if there is an upper limit after which a battery may fry the inverter.
>>
File: fotd.png (19 KB, 600x600)
19 KB
19 KB PNG
filter of the day
>>
>>1304533
the browser? sadly, no. gotta eyeball it with grid lines or load the footprint up in the editor and measure there

>>1304585
>more difficult
not really
>more expensive
yes because car batteries are literally not designed to be deeply discharged. golf cart batteries and such are, and they'd be okay for this application
>LED bulbs
show 'em the amazon chink strips
>phone chargers
buy a bucket of https://www.aliexpress.com/item/Fine-6-24V-12V-24V-to-5V-3A-CAR-USB-Charger-Module-DC-Buck-step-down/32757316311.html
>laptop charger
you got me there, but undoubtedly aliexpress has a way to run that too from 12V
>upper limit on battery
more likely a lower limit on battery, past which the inverter will over-discharge the battery or discharge it too quickly and fry it. they really should skip the inverter.

>>1304586
>gimmick inductor
I keked
>>
>>1304585
>I just want to be sure that things don't go kaboom or if there is an upper limit after which a battery may fry the inverter.
The inverter only cares about the battery's voltage, not its capacity.
So there is no upper limit.
Anyway I don't know if gel car batteries are any better for deep cycling. Maybe.
If someone charges the battery every week then maybe discharging the battery will never be a problem, go for it.
>>
>>1304589
Ok I get your point, I will try to see if I can find cheap golf car batteries then, but if they turn out to be more expensive than several months of grid power, then I'll have to fall back to car batteries and call it a day, I just want to make sure safety is not a concern and that if something has to fail, then it better be the battery and not the inverter or all the utilities attached to it.

What's that thing you linked me? Sorry if it's a stupid question, but again I'm very out of the loop when it comes to electronics. Is this a thing that I can trust a group of 70+ years olds to safely set up and use?

Also how do I figure out that limit on the inverter? Is there a specific value about the inverter that I can use to calculate such a limit?

Also, about chink strips, they could be a good idea, but the wiring is already there and ready to use, everything is already set up and functioning at 220V with wall outlets, switches and everything, they would like to still have the convenience of using the same hardware they have now without rewiring every bulb and outlet to install stuff rated for 12V.

Sorry for all these questions, I want to make sure I can buy the right stuff and let them have their fun so they don't have to call me every week because they don't know how to fix the things they will inevitably break.
Thanks m8

>>1304595
Ok, notes, thanks a lot to you too
>>
File: 147392506748.png (189 KB, 640x475)
189 KB
189 KB PNG
>>1303811
Hi.
A few weeks ago, I posted here to request some help to make a SRAM tester for an old Tektronix VM700A which has a problem with the audio processor board.

Well, I finnally made the code, I found the bad SRAM chip, I replaced it, and the error dissapeared, but hey, that would be too easy.

Previously, before "repair" it, it showed an error from audio RAM, and when it ran a few minutes, it restarted itself. Well, with the new RAM, the RAM test is OK, no problems, but when it is running a few minutes, it restart itself. Wow.

Instead of crying like a baby, I probe to find the problem. One of the boards has some switches, and one of them cancel the auto reset (that's how Tektronix call it). Yeah, that's great.
>Tektronix: No.
>Me: Why?
(Crying.exe)
I supposed that the problem might be in the audio processor board, (Without ths board, the machine does not restart itself).
And well, I have no documentation, schematic, only the pin in the bus called SYSRESET. Follow the wire, and it goes to a hundred IC's (ok no, Only 3 ICs, and some of them are programmable arrays, and I don't have the code for them)
Well, let's find the fail in software.

If you try to make an audio test, the machine (Levine) freezes, not restart, only freezes, and I need to shut it down. Lets find the directories of tests (They are like exes).
>Audio_Measurement
>Directory doesn't exist.
>...
>Same with all audio test.

Good.

Somebody deleetd audio tests, or they magically deleted itself from the Non-Volatile RAM.

Well, I'll try to re-install the card via PC.
Oh, wait, I have tried it. You need the CD and the key taht comes with de Expansion card (In this case, the Audio Processor and the audio analog board).
Well, I don't have it, maybe the codes are writted in the board.
>It could be.
OK, I'll try to install it, I'll download the CD from the Tektronix website.

>Doesn't exist.

I wanna die.

I'm not here to request help, only laugh to my situation.
>>
I took apart a massage thing and inside was a toy dc motor and a couple of LEDs. Is there a good resource for me to use so I can identify the intended voltage the motor is supposed to be run at?
I took it apart because the previous batteries in there burst and it stopped working. I think the corrosion made the batteries unable to contact the springs properly, but after taking it apart and contacting a battery directly to the motor wires, it budged a little but did not rotate.
>>
>>1304620
if your sex toy doesn't have any other wound coils in it then that motor should run off the batteries alone. you say it has "batteries" and then test it with "a battery" though.
>>
I've decided to learn to use eagle. Right now im looking for a nice resource to get myself started. Looking to make backups of my projects basic schematics and layouts to organize myself a little better, and design the pcbs i need in the future.

The thing im noticing is that it's hard to find the components I need.
How do you do it, diy? Do you know any good site to get libraries? Or do you just make them on the fly?

Know a good reference, a good book or something? Maybe the documentation and manuals in autodesk is enough?

It's been a pretty good day, hope you are all having a good one aswell
>>
>>1304598
>that thing
12V-5V converters. if they want chargers, they can just pull directly off the 12V battery through one of them. you/they could just wire a bunch of them to the battery. throw some insulating heat shrink around them, attach a stack of micro USB cables, and that's all they need. they might even work better and charge faster than their wall warts. more power efficient, at the very least.
>limit
A drawn on low side ~= A drawn on high side * (20 + a bit)
note that lead acid batteries are generally rated to be drawn down over 12h. if you have a 110Ah battery, that translates into ~5Ah of 240Vac, minus a bit for inverter losses. the faster you draw it down, the less usable capacity, and in extremis, the less service life of the battery. you might need to buy multiple batteries and parallel them if their current requirements are much greater than that, which makes it a less profitable exercise.
problem is, "a bit for inverter losses" could be a lot. inverters will draw more than 100% of the power drawn from them and turn it into heat. they will also draw some amount of current when idle. you can measure the latter easily by connecting the inverter with no load to a stiff power source like a running car with a good alternator, in series with a power ammeter (like 30A or more full scale, maybe your boss has one).
>wiring is already there
I hope you mean extension cords and power strips. back-feeding power into permanent building wiring is unwise.

>>1304633
there are really a LOT of components out there. for any EDA software, making your own components is fairly routine, footprints only slightly less so.
aside, I found myself exceeding the EAGLE free tier's limits so I gave KiCAD a try. it's a much better program than it was 10 years ago, for sure, and I've gotten some useful work done in it.
cheers to ya
>>
File: IMG_20171229_140801.jpg (2.88 MB, 4032x3024)
2.88 MB
2.88 MB JPG
>>1304632
I don't think it was intended as a sex toy but I guess if you were determined. The entire thing was powered by 3 AAA batteries in series, and the reason I thought the problem was with the battery holder was after putting the batteries in, my voltmeter read about 0.4 V while individually each battery read 1.59V. Last night I tried just 1 AAA battery, the motor did not rotate.
Tried it again, 1 AAA battery works, but it rotates more vigorously with 2 in series.
I think my question is how can I figure out the intended voltage for it not to burn out since it does not have markings on it for me to find a spec sheet.
>>
>>1304650
>I hope you mean extension cords and power strips. back-feeding power into permanent building wiring is unwise.

I mean they will put the inverter where the grid power meter (if that's how it's called in English, I mean the main switch that powers off the whole building and calculates how much juice you use so you can be billed later) is now. Since they plan on cancelling the grid power supply, they will remove the meter and install the inverter on its place so they can then keep on using light switches and wall outlets like they are used to do. I have no idea if that's a safe thing to do.

Thanks for all the other info
>>
>>1304710
as long as they're wiring it in properly, i.e. they're not using a cord with a Schuko plug on each end, at least nobody's going to kick a live plug out of the wall and electrocute themselves. should be alright, then.
cheers
>>
>>1304674
then you can safely assume the motor is rated for x torque and y rpm at 4.5V and you could probably push it to 5 without blowing it up. note the diode, maybe you just hooked it up backwards last night?
>>
>>1304674
>The entire thing was powered by 3 AAA batteries in series,
>>1304674
>how can I figure out the intended voltage
4.5vdc
>>
File: image.jpg (10 KB, 300x135)
10 KB
10 KB JPG
What type of capacitor should I use on the input to a VR?

The datasheet says it's necessary (or at least helpful) if the VR is far from the DC supply, my cheapo Bluetooth MP3 player thingy didn't come with one, and seeing as I'm using a long power lead and replacing the VR (which failed short) and the output capacitor (which went on fire), I thought I might add one.

Voltage and capacity is in the datasheet, but chemistry isn't. I'm guessing electrolytic isn't correct, but what would be?

Also can I dead-bug it across the top of the VR package?
>>
>>1304764
>what kind
on the output, the type of cap matters a lot for stability. but the type matters less on the input than having enough. several uF is usually enough, and it's a good idea for the input cap to be larger than the output cap. ceramic or Al-el should serve just fine. dead-bug is fine. even better if you're a badass and you slap an 0805 on the leads.
BUT
if it uses a USB voltage input and not a barrel jack, you should probably not exceed 10uF on the input.
>>
>>1304764
>78Mxx
Anything goes, but if you're going to use the values your pic suggests, ceramic and plastic capacitors are the usual choices.
You can solder them on top of the regulator.
>>
>>1304773
>>1304769
Cheers for the help!
>>
Hey guys, I bought this amplifier and I was wondering how loud I could push a 15W 8ohm speaker with it. I tried connecting the aux cord as a quick test and with 6W of power going in it was REALLY quiet.

https://www.ebay.com/itm/1PC-TDA7492P-50W-50W-Bluetooth-4-0-Audio-Receiver-Digital-Amplifier-Board-US/132163798077?hash=item1ec593843d:g:kIMAAOSwi8VZSl7i

It's advertised at 50+50W so does that mean I can put a lot of current through this? This is my first time using the circuit so I'm kind of clueless.
>>
>>1304457
Much less than 0.1A, and it should be more like 400V.
>>
>>1304747
>>1304757
I suppose that makes sense. Since there were some LEDs in the circuit I assumed they would be involved in the voltage drop. Looking back at the picture I took when disassembling it, you guys are right: the LED portion was hooked up in parallel to the motor meaning it got the full 4.5 V. I appreciate the help.

I tried the battery in both orientations and the motor simply switched direction of rotation. I think I was just having a hard time getting the leads to contact thanks to the corrosion. I think the diode is there to prevent back current from motor reaching the LEDs when shutting off.
>>
>>1304790
open schematic
show wiring
>>
We got some cabinet heaters at work today. You had to supply the wiring. On the unit where you hooked up the power there were ports labeled L,N1, and N2. What is N2 supposed to be used for? I've never seen any piece of equipment with 2 neutral spots. It didn't come with any instructions and I couldn't find anything on google related to that. I just wired it with line in on L and neutral on N1 and it worked but it still has me puzzled.
>>
>>1304813
didn't come with one :(

I'm going to try and find one online.
>>
>>1304790
>I tried connecting the aux cord as a quick test and with 6W of power going in it was REALLY quiet.
Is the aux an input or an output?
Going in what, exactly? Your speaker?
How do you know it was 6W of power?
If the aux is output, it's probably signal level output, not amplified, which would explain why it's so quiet.
I built an audio amp from online schematics once and it was max 6W output per channel and that was plenty loud enough to blast tunes in a small music room with a set of loudspeakers.
Hook up your speaker to the terminal blocks instead and try again.
>>
>>1303811
when shopping for an oscilloscope, which qualities are absolute must-haves, which are priorities or preferential, which are nice features, and which are stay-away?
what kind of price range can I expect to be looking at?
>>
>>1304900
$50-100 for used 2 channel analog scopes that mostly work fine. $350 is the lowest digital scope i know of. $400 for the 4 channel rigol ds1054z that everyone has. it really just comes down to budget, not feature count, unless you want to drop more than $400.
>>
>>1304903
are there any applications where a digital scope works and analog doesn't? or the other way round?
>>
File: PID.png (45 KB, 1889x927)
45 KB
45 KB PNG
Im trying to simulate a voltage-mode type III compensator PID op amp (see pic) to regulate a buck converter using LTSpice.

I have derived the PID transfer function using MATLAB so that the open loop transfer function of the buck converter has a phase margin of 60deg. I am following guidelines from a texas instruments applications note on how to choose the R and C values of the op amp compensator but I still do not achieve the correct transfer function response of the PID op amp.

Does anyone have any experience with this?
>>
>>1304908
Digital scopes (can) have features that let you precisely tell the RMS value, average, and specific voltage and time intervals as you move about the cursor. Analogue scopes can have the last feature(s), but are typically used where precision of that nature is not important. You can also buy cheaper computer oscilloscopes, but I wouldn't.
>>
>>1304938
thanks
i will keep researching, find my budget, and get a scope that is worth it
>>
>>1304908
Digital: if your event is so rare that you can't see it with a (normal) analog scope.
Digital: if you want to process the measurement data with a computer.
Digital: if you want to take lots of nice-looking screenshots.
Digital: if you want the scope to perform some simple math with the waveform, like measure frequency, amplitude and so on.

Analog: if your signals have so wide bandwidth that the aliasing effects are a problem.
Analog: if you have low budget.
>>
File: wtf.jpg (31 KB, 512x358)
31 KB
31 KB JPG
>>1304917
R3 is less than 1 ohm, C2 is close to 20 uF
WTF is going on
>>
Time to buy a new soldering iron station. Recommend something below 100 usd. Im going to be using it fairly regularly
>>
>>1305101

Excellent answer as I expect when I see your name.
>>
>>1305134
>>1305101
> oh boy here we go.
>>
File: psu.jpg (36 KB, 1152x648)
36 KB
36 KB JPG
Rate my psu.
>>
>>1305102
Aoyue 888A, from sra-solder. hot air and 60W pencil for $75, ships from east coast..
>>
>>1305153
The second capacitor isn't necessary bro
>>
>>1305153
floating short psu
>>
>>1305134

Thanks Mr. Bill. We need to support each other and ignore those who shun us.
>>
>>1305176
Fruitcakes!!
>>
File: 1483187160042.jpg (12 KB, 480x360)
12 KB
12 KB JPG
>>1305178
actual photo
>>
>>1305152
Well it's autism at its finest. Flooding, spamming and generally acting plain weird.

I've seen few autists (two, to be precise) in real life. Poor people. Without internet (and computers) they would likely loose a major way of communication with society and loose their social skills fast. You gotta be patient...
>>
>>1305102
>>1305161
Can someone suggest an alternative for Europe?
>>
File: 1511371571839.jpg (64 KB, 565x463)
64 KB
64 KB JPG
Hello folks,

I added a small fan to the side of my PC case. Problem is, it keeps going at 4500 RPM and can't be regulated automatically (motherboard limitations).

What I want to do is put a potentiometer on the fan and regulate the RPMs manually when needed.
What kind of resistance should I look for in the potentiometer? How do I decide this?

It's my first small project getting into ohms'n'shit.
>>
File: asdfasd.png (24 KB, 413x481)
24 KB
24 KB PNG
>>1305183
>>1305134
>>1305101
Don't forget to filter it anons. He's a samefagging cunt here for the e-fame, nothing more, and if you reply to it, you're just as much of a cunt for feeding it the attention it so clearly desires.
>>
>>1305207
a couple of threads ago >>1295428 someone needed much the same thing. might review the ideas presented there. or search the web for fan speed control circuits for even more possibilities.
>>
>>1305220
You must either have reading comprehension issues or mentally challenged if you think I have something to do with another person who is replying and talking to himself.
>>
File: IMG_20171230_155445.jpg (2.72 MB, 3968x2976)
2.72 MB
2.72 MB JPG
Anyone know where I can get a switch to go into this shitty panel?
Need a SPST rocker that can handle about 4 amps at 250V AC.
Cutout is only 6mm by about 19mm
>>
>>1305238
...did you cut that out without having picked a switch?

if you need it now, troll thru the walmart equivalent or auto parts store and find one that fits and read the print on the side of the switch itself to get the real rating
>>
File: NKK CWT12AAS1.png (102 KB, 1599x741)
102 KB
102 KB PNG
>>1305238
maybe you can make pic related fit
>>
>>1305245
damn, looked at the wrong spec. this one only handles 3 A
>>
>>1305242
It's an off the shelf panel.
Suppose I could take the dremel to it to make something fit
>>
>>1305176
>>1305134
You forgot to switch your trip shitlord.
>>
File: fan-pot-1.jpg (9 KB, 500x213)
9 KB
9 KB JPG
>>1305207
500Ω 2W
>>
>>1305207
First lesson of "ohms and shit":
Potentiometers are not ideal dimmers.
>>
>>1305397
>>1305411
Thank you.
>>
>>1303821
only 179 until we can start on the ubiquitous cmos series
>>
>>1304536
looks like audio cable to me.
but everything does i suppose.
>>
>>1305207
how big is your fan? (voltage/current)
some sort of current limiting or maybe even PWM is better than just using a pot as a variable shunt resistor.

could be a good learning project.
>>
>>1305448
at this rate this thread will probably reach bump limit sometime in 8xxx and damned if I'm going to post some crappy intel 8-bit periphs from yesteryear
>>
>>1305454
Maybe if we're lucky it will land on a 74xx.
>>
>>1305229
> he thinks we are as retarded as he is.
IP counter doesn't lie you bipolar fuck.
>>
File: IMG_20171230_235356.jpg (26 KB, 894x671)
26 KB
26 KB JPG
haha fuck these SMT chips are a lot smaller than I anticipated, I don't know how I'm gonna solder it
>>
>>1305624
They're not that bad. Flux + sensibly sized chisel tip iron and you'll be all good.
>>
>>1303811
Maybe this would be better for next OP.
>>
File: IMG_6727.jpg (3.58 MB, 3024x4032)
3.58 MB
3.58 MB JPG
So I'm trying to make a little board that will fire an ignition coil through a spark plug. Short story, I'm going to build 6 of these and integrate a driver board for my spark plugs in an engine management system.

This is the circuit that I have come up with. The ignition coil is (as most are) pretty much just a step up transformer.

I have a couple problems.
1. The MOSFET triggers, but there's like 2 volts across drain and source, even with 5V on the gate. I tried 2 different MOSFETs that are both supposed to be logic level threshold voltages, but neither one seems to want to saturate. I know that an IGBT would be better, but I need a test rig right now and I'm working with what I have. Is there some way I can drive an N-channel MOSFET with BJTs?

2. My power supply is clearly not fit for the job. I have a 12 volt supply, but it's only rated for 1 amp. The primary coil of the coils that I'm testing is .8 ohms and you're supposed to drive them with 12 volts. This means I need 15 amps. What's the cheapest way to get a 12V, 15A supply? Harbor Freight sells a 10 Ah SLA battery, but it's $40. I could theoretically use my car battery, but that might start to get annoying if I have to go pull my battery out of my car every time I want to use this test board. Any ideas?

I'm fine with running a battery at 2-3C because the pulses are going to be 5 milliseconds maximum. I've tried using some large capacitors but it doesn't seem to be enough.

For the record, I've tried just brushing the primary coil across a ground contact really fast and it trips my power supply 100% of the time. Doesn't do anything when I trigger it with the MOSFET though.

Ideas?
>>
>>1305668
this is a bad idea
>>
>>1305669
Do you have a better idea?
>>
>>1305668
Use a 2-transistor voltage follower on the input side?
>>
>>1305668
>>1305671
The arduino is not engineered for automotive environments if that's what you're doing.
Google megasquirt or diyefi (open source engine control projects) to research the on work already done for engine computers.
You might as well study what's been done before.

1. google BJT gate driver circuit, or use dedicated gate driver ICs. gate drivers should be able to supply high current (for both charging and discharging the gate) so that switching times are low. using a microcontroller gpio pin for driving mosfet gates directly can be a bad idea.
also you should protect the mosfet from inductive kickback. google ignition coil snubber, there are multiple ways to do this.

2.
>Harbor Freight sells a 10 Ah SLA battery, but it's $40.
$40 is not a lot to spend especially on a project like this, and no decent quality powersupply will be that cheap.
your pc's atx power supply is going to be cheapest, because you might already have one. if not call around for a used one. used computer stores have boxes full of that kind of shit.
or ebay search for a chinkshit powersupply with your desired specs. might be complete garbage tho.
>>
>>1305682
Oh no of course! I have an MS3. I'm just using the Arduino as a test signal generator to make sure whatever coil driver I'm using works, and that the coil works. I didn't mean to seem rude.


1. I put a 20V zener on it just for now. I'll find a better way to do this in the final installation. Again, this is just a test rig (and honestly I'm kinda screwing around a little). I will look into dedicated driver ICs too. Didn't know those existed.

2.By the time I bought an ATX supply, I might as well just buy a battery.

I guess I should just spend $40 on a battery.
>>
>>1305668
You need a high voltage FET or IGBT and those invariantly want >= 10V gate drive.
>>
>>1305624
tack down one pad to hold position, flux, work your way around. use a light touch, a small*ish* screwdriver tip, and thin solder wire (I like .020").

>>1305707
lead acid batteries don't like to be drawn down at 2-3C. might be better off buying a marginal one from a recycler
>>
File: 2018.png (12 KB, 240x151)
12 KB
12 KB PNG
TLE2426
>>
>>1305668
You should bypass the FET with a capacitor to ground. There's a reason why old cars had one. It absorbs the inductive kickback across the primary, and then releases it, making the primary "ring" for a short period. This makes the spark stronger.
Probably .2 - 1uf.
>>
>>1305668

>The MOSFET triggers, but there's like 2 volts across drain and source
The max gate threshold voltage is 2.5 V, so mcu output should be enough. Are both sharing the same ground?
>>
>>1305802
Yep, same ground. All running off the same power supply.

I just woke up and I'm going to keep fiddling with it.
>>
File: IMG_6728.jpg (927 KB, 4032x3024)
927 KB
927 KB JPG
Somehow, it works now! The spark is pretty weak, and the PSU starts to drop voltage when you start to pulse it too fast, but holy shit it works. the MOSFET starts to heat up too. The PSU is just a crappy wall wart though, so I'm not too torn up if it gets shrekd.

As soon as I took the Zener off it started ticking away. I replaced it with a capacitor per >>1305793 since I would hate to kill my microcontroller or murder my FET.

I guess the main point of this exercise for me was knowing how these coils are hooked up and how to fire them. I got 8 of them from the junkyard expecting to torch 2 of them and have 6 left. But now I just have 2 spares!

Thanks for the help guys!
>>
>>1305668
needs a reverse diode across the pimary to protect the transistor.
transients will kill it.
>>
>>1305849
No. That kills the spark.
You can use some limited overvoltage protection, but you also need a high voltage transistor. Typically transistors intended for ignition coil drive can handle several hundred volts.
>>
What's the highest voltage anyone here has used in a breadboard?
>>
>>1306060
180. ipc suggests 300 would be the max before arcing became a concern.
>>
>>1306060
solderful or solderless?
>>
>>1306063
Well I'll be putting 400 through one, I guess I should leave an extra gap or two between each high voltage row and the others. It's also the cheapest breadboard I could find from Ali, so I might be in for a shock.

If this proves to be a problem I'll probably just solder the HV parts of the circuit together in mid-air, it's only 6 components or so.

>>1306071
Solderless.
>>
>>1306090
I recommend air wiring if you're not up for fabricating a pc board. recommended: solder in a neon bulb with resistor so that you know not to stick your fingers anywhere near it
>>
>>1306093
Well I did make a PCB and soldered everything to it the first time. But it didn't work. This thing can't deliver any current so I should be fine putting my fingers around it. It's DC anyway.
>>
>>1306100
it only takes a few mA discharge for a short period to kill you. keep a hand behind your back, figuratively or literally.
>>
File: boost.jpg (190 KB, 1200x800)
190 KB
190 KB JPG
>>1306060
i had no problems with my nigger rigged 400 V boost converter
>>
>>1306107
Close enough! What is your boost converter for anyhow?
>>
File: meter.jpg (770 KB, 1500x997)
770 KB
770 KB JPG
>>1306114
was merely testing what a 555, choke ballast and mosfet could do. 400 V was the limit of the diode i used.
unfortunately a critical resistor array appears to have died in my 80s bench meter during the experiment for which i have not yet found a replacement
>>
>>1306120
Well good luck with that. Got a part number for me? There's an old surplus electronics store where I live and I might be able to see if they have one.
>>
File: R.jpg (371 KB, 1500x997)
371 KB
371 KB JPG
>>1306130
it's this fucker right there
100-268Q 8808
>>
>>1306120

is that a early 90s fluke?

it maybe easier to just buy a wrecked one
>>
>>1306139
it's an Hameg HM8011-3.
the resistance and current ranges still work and i've already bought a new bench meter so it isn't that problematic to me.
finding a value-matching resistor array isn't actually that difficult, it's just that none of the available footprints match. i might just fiddle one with some wires to the board at some point
>>
>>1306134
There are a few single-inline packages there, I'll have a look next time I'm over.
>>
File: HM8011-input.png (170 KB, 759x685)
170 KB
170 KB PNG
>>1306134
10MΩ precision divider, standard values.
>>
>>1306452
hmm, Vishay CNS471A6 looks like it would fit, but that one would set me back by 60 €.
I think i'll buy a Caddock 1776-C681 for 16 € and simply fix the pin layout by hand.
>>
tldr: is an hc-05 bluetooth module not able to enter AT mode while connected via bluetooth?

long version: Mine has a button that you press before applying power to enter AT mode, which is confirmed by the led blink rate changing from 2hz to 1hz.

However that is not very convenient. The pdf (chinese-english) seems to say that at any time you can raise the Key/Enable line and enter AT mode, which is what the pushbutton does.

But I want to do that while it's connected via bluetooth to an android device. When I try that, using an AVR to pull the line high, the led still blinks the "connected" sequence but no longer responds to bluetooth packets and ignores AT commands from the AVR UART as well. If I reset the AVR so that it lets the line drop, the HC-05 is still connected to the android tablet and resumes communicating.

Incidentally, this is one huge advantage the Sparkfun bluesmirf has in that it can enter AT mode via bluetooth or the UART, while hc-05 can only accept AT via the UART. But the smirf costs about 10x more so I'm trying to get the hc-05 to cooperate.
>>
>>1306465
>>1306452
Why not assemble this from discrete resistors? does it need better than 1% precision?
>>
>>1306542
the ratio is matched much better when they're fabricated together.
>>
>>1306542
It largely determines the accuracy of the voltage range. 1% accurate multimeter is Harbor Freight tier and not something you'd expect from a benchtop instrument.
>>
>>1306570
The gap is fine. It's better than the box sticking out too much.
Try for the best fit you can when cutting around the boxes.
>>
>>1306588

if the screws are long enough to engage several threads it will be fine. the cover plate is supposed to fit firmly against the sheet rock so they did the right thing by not having the box out too far.
>>
>>1306598
Cool, thanks for the help. Moved my question over here. >>1306578
>>
File: file.png (12 KB, 190x301)
12 KB
12 KB PNG
>>1306810
>>1306813
the body diode in mosfets isn't purposefully added, it's just a byproduct of the process. it won't help you clamp transients here. what you want is a snubber, as pictured. if you have info on your transformer's inductance and turn count i can show you how to calculate the values. otherwise you can just ballpark it.
>>
File: IMG_6744.jpg (3.39 MB, 4032x3024)
3.39 MB
3.39 MB JPG
I'm the guy who's trying to build an ignition coil driver for a standalone ECU install.

So I'm kinda new to this and I want to make sure I'm not going to explode anything, because this is going into a car and it needs to be reliable because it's a life safety issue.

The MOSFET I'm using is an NTE2987 (or the equivalent to an IRL520). The datasheet has a reverse diode in the schematic symbol, or what I'm pretty sure is referred to as a source-drain diode. I'm assuming that this is included to clamp inductive transient spikes, but I don't know this for sure and you people are smarter than I am, so here I am.

Pic is the scope trace of my gate signal (square wave) and the drain (or the secondary of the coil). There's still a pretty big transient spike, but if you zoom in...
>cont'd
>>
File: IMG_6745.jpg (3.27 MB, 4032x3024)
3.27 MB
3.27 MB JPG
>>1306816
This is what the trace looks like at the drain. The spike hits a flat spot at ~100V. Does this look like what it would look like if there was a clamping diode on there?

I also have a 1N4001 that I tried, which is supposed to break down at 50 volts. Putting the cathode on the drain and the anode on ground does nothing. Doing this with a 20 volt Zener diode kills the spark.

Is this going to murder my FET?
>>
>>1306819
i'm of the opinion that the 100v cap there is just a great coincidence. when you turn off your fet you'll see Vin+Vout*n+(leakage inductance discharge) at the drain. i don't know why your fet is still alive with what looks like a 115v transient on it but process variations and especially temperature can result in a higher than specified breakdown voltage. theoretically the leakage inductance on your transformer should destroy your fet because it lacks a discharge path but it's possible the fet's drain source capacitance is absorbing the leakage energy. i think you only managed to get this to work through luck.
>>
>>1306822
>luck
Yeah that sounds about right.

What's with the flat spot on the spike though? It seems like that's exactly what a clamped spike would look like to me.

If my FET was going to release the blue ghost, how fast would that happen? Would it take 1 pulse or 10 or 10,000? I've left it running at like 5 Hz for the last 2 or 3 hours and it's still going.

Why does a 20 volt Zener between the drain and ground kill the spark? Why does doing the same thing with a 1N4001 do nothing at all?
>>
>>1306815
I have no idea what the inductance of the transformer is, unfortunately. It's an ignition coil off of a BMW from 1995. The primary is .8 ohms though.

What would be some good values to start with for that diagram?
>>
>>1306831
i'm not knowledgeable about spark gaps, but i'll speak generally about flyback converters like yours. when you shut off the fet, the current built up in your transformer will continue to flow until the magnetic energy in the core is dissipated. as it fills up tiny parasitic capacitances it'll quickly increase the voltage on any windings until it finds a new path for that current. the actual voltage it creates is entirely arbitrary but the voltages on the two coils are always scaled according to the turns ratio. so as that magnetic dissipates you can get a very high voltage on your secondary coil when it finally arcs while only having a lower voltage on the primary. eventually it will either trigger your spark plug or it'll cause arcing in your fet, destroying it. because the discharge occurs over time that voltage will be held until the energy is dissipated. then you'll get that ugly ringing until the fet turns on again. that flat spot is clamped in a sense by your plug.

adding a 20v zener obviously gives the magnetic energy somewhere else to dissipate before the voltage ever gets high enough to cause your plug to arc. adding the 1n4001 should have the same effect but reverse breakdowns on diodes aren't consistent unless they're designed to be.

if the magnetic energy had to discharge through your fet then it would've died the first time you turned it off. given that it works i'll assume your leakage inductance (basically a regular inductor in series with your transformer. it can't discharge through the secondary.) is low and is being sufficiently absorbed by parasitic capacitances so i wouldn't bother with the snubber with your current configuration. i'd definitely get a better fet though and snub that because transistor lifespan decreases geometrically as you approach breakdown or junction temperature limits.
>>
>>1306839
>i'd definitely get a better fet though and snub that because transistor lifespan decreases geometrically as you approach breakdown or junction temperature limits.
This is what I'm afraid of. Having a FET fail while driving the car wouldn't be fun. I'm looking at IGBTs that are designed to drive coils like this for the final design, but for now I'm trying to gain a good understanding of what's going on.

So it goes like this:
>the FET switches on
>current in the primary increases until the transformer core reaches magnetic saturation
>the FET switches off
>the magnetic field in the primary begins to collapse
>the secondary absorbs some of this energy and the spark plug fires
>the magnetic field collapses to a point where the voltage in the secondary isn't enough to arc across the spark gap
>field keeps collapsing, driving current through the primary and creates the transient spike seen on my scope

Kind of? Or the spark and the transient happen at the same time?

I guess I'll start reading about snubbers.
>>
File: file.png (36 KB, 477x346)
36 KB
36 KB PNG
>>1306842
not quite. see picture.
1) fet turns on and current starts flowing through the primary
3) fet turns off, stored magnetic energy results in a voltage spike until the energy finds a discharge path. the current, in yellow, is flowing through the plug because that's the first thing to arc. because of the turns ratio it'll be a lower current and higher voltage. i drew it badly. it stays at that high voltage until nearly all of the energy is dissipated, because the voltage is just high enough to cause arcing. seems kind of weird that the arc voltage would stay so steady throughout the discharge but i don't know spark gaps.
2) fet turns off, leakage inductance (inductance that isn't coupled to the secondary) also needs to discharge its energy. because it can't go through the plug on the secondary side it just discharges into parasitic capacitances. you're just lucky that those capacitances don't build up enough of a voltage that your fet breaks.
4) LC circuit created by the coils and parasitic capacitances will oscillate and emit electromagnetic infetterance with the small amount of residual energy. you can't do anything about this in your configuration. it's a characteristic of "discontinuous current" converters which yours will always be. on the primary side this oscillation is kept from going negative by the fet's body diode.
>>
>>1306848
So if the stored magnetic energy is discharging through the spark plug, then the FET will immediately become kill if it was removed. Giving the energy nowhere else to go would not limit the transient spike at the FET drain, overvolt it, punch a hole through the gate layer, fail closed, melt my coil, and kill itself.

And if I put in a narrower gap, the spike will be "clamped" at a lower voltage? Not planning on using this to snub it, but logically it should if this is what's going on.

Thanks for the graph bro.
>>
>>1306819
The flat spot looks like the drain-source breakdown voltage of that mosfet. I've seen exactly that effect when I was dabbling with making a boost converter.
Interestingly, upon some quick googling, I found application notes on using ignition IGBTs but did not see anything for ignition mosfets. Google ignition igbt and you'll find a bunch of parts.
If you're just playing around trying to understand this stuff then yea a mosfet will do the job.
>>
>>1306869
Wait hold on though, if the flat spot is from the D-S breaking down, why is the FET still working?
>>
>>1306871
and additionally, why would the plug arc if the voltage is being clamped by the fet rather than the plug?
>>
>>1306815
Ok, so I rigged this up with some of the capacitors I had on hand (just the little ceramic disc ones), a 1N4001, and a 5k potentiometer. It works pretty well. Most of the gross high frequency stuff is gone.

A couple of things that I noticed:
-If you trim the resistor low enough, there is no spark. I'm guessing that this is because the path of discharge from the primary collapsing is easier through the resistor than it is through the secondary.
-More capacitance allows you to trim the resistor further down before the spark stops. I'm not sure why this is.
-Applying the snubber weakens the spark. I wired it to a tactile switch so that I wasn't hot swapping components, and you can definitely tell that it's weaker when the snubber is applied. Is this necessarily true? Can you suppress the transients without weakening the pulse on the secondary? I would think not, because no matter what you do, the path through the diode and the resistor is providing an alternative discharge path that will rob some energy.

This is really interesting.
>>
>>1306878
here's some reading material
https://www.vishay.com/docs/90160/an1005.pdf
reaching avalanche breakdown voltage doesn't necessarily mean your mosfet will be fried.
not all mosfets are constructed alike, and some are more tolerant than others of sustaining pulses at the breakdown voltage.

>>1306878
Why wouldn't it? Turns ratio on an ignition coil is huge.
>>
Is the voltage spike on the primary roughly n1/n2 times smaller than the voltage spike in the secondary?
>>
>>1306871
Avalanche breakdown doesn't instantly kill the body diode, some mosfets are actually rated for it to be okay. But really, you should be providing a proper back EMF path through a proper diode.
>>
>>1306997
Bulk of it is, but you can expect a part it being higher due to leakage and wiring inductances.
Typical ignition coil turns ratio is something like 1:100, so 100V at primary isn't that much.
>>
>>1306819
>Is this going to murder my FET?

Nope if you are within absolute maximum rating of "Ear" or "repetitive avalanche energy" for your MOSFET. Basically your ignition circuit is 1-to-1 copy if UIS test circuit.

>>1306822
>i'm of the opinion that the 100v cap there is just a great coincidence.

Body diode in MOSFET's is actually a zener, for a 100V mosfet it's a 100V zener, for 200V MOSFET it's a 200V zener. And it's not a limp one, it can dissipate about the same amount of power as a mosfet can. Again, check data sheet. Maximum avalanche current for IRF520 is 9.2A.

>>1306831
>If my FET was going to release the blue ghost, how fast would that happen? Would it take 1 pulse or 10 or 10,000? I've left it running at like 5 Hz for the last 2 or 3 hours and it's still going.

It's safe. Basically if the FET isn't too hot and avalanche current is less than 9A you are good to go.

>>1306831
>Why does a 20 volt Zener between the drain and ground kill the spark?
You are clamping voltage across primary winding and since primary and secondary windings on igntion coils are magnetically coupled you clamp output voltage as well.

>>1306831
>Why does doing the same thing with a 1N4001 do nothing at all?
It's not a 50v Zener, it is designed to withstand at least 50v or whatever number but there is no guarantee it will break down at that voltage.

You can test break down voltage safely: you need high voltage source and like 1-10 Megohm resistor in series. Then you rise HV until you register 1-5-100uA (or whatever number the datasheet specifies).
>>
>>1306835
>What would be some good values to start with for that diagram?

There is numerous IGBT's designed specifically to drive ignition coils (some are smart enough to have over-current protection). I'd suggest you to switch to 350-400V devices and/or copy ignition drivers found in VEMS/microsquirt/megasquirt/rusEFI/etc.

BTW why don't you go with COP's with builtin drivers? For example you can get COP's from yaris for cheap and they are quite successfully used on 2JZ.
>>
>>1307015
Ok, so the little symbol in the datasheet that looks like a diode going from source to drain is actually notating a Zener diode, and the Zener voltage is roughly the maximum voltage that the FET is rated for. It breaks down and conducts if the potential is more than this value, but dissipating more than the "maximum repetitive avalanche energy" (which is 30 millijoules in this case) in a pulsed application will kill the FET.

So two things are happening. Parasitic capacitances in the coil and the FET itself are absorbing some of that energy, and the energy actually passing through the FET must be lower than the Ear rating.

>You are clamping voltage across primary winding and since primary and secondary windings on igntion coils are magnetically coupled you clamp output voltage as well.

This explains why clamping the spike with a snubber significantly weakens the spark too. That big nasty spike is required in order to have a high voltage in the secondary?

And for that reason, what I really need is a high voltage tolerant, logic level switching device. And then use a snubber not to completely kill the transient, but only to dampen the high frequency LC ringing so there isn't a massive amount of RFI under my hood fucking up everything.

>>1307020
>2JZ
>self driving coils
>Yaris
Nah. I'm something of a purist, and the engine is a BMW M20. I'm insistent on using BMW coils, and I got 8 really clean ones from the junkyard the other day. It would be way easier, yeah, but just that little bit less cool. And I probably should go over to a more specialized forum.

And yeah, as I said, I'll definitely be using an IGBT that's intended for this exact purpose in the final design. I'm just trying to understand the fundamentals of what's going on. This stuff is all fascinating.

Thanks everyone for all the help. Tonight I'll read about bespoke ignition coil IGBTs and order a set. I'll post more when I've built something.
>>
>>1304558
I'm thinking it is coax for a sat phone antenna. Are there any markings on the cable jacket?
>>
sup /ohm/. im trying to fix my mom's space heater. i found a blown GUERTE TR-1506 rectifier diode rated at 600V 15A. quite a beefy fellow, this space heater is kinda old like maybe 6-8 years.

anyways, i have these 1N4003 rectifier diodes that are rated for 200V 30A. i was thinking of replacing the TR-1506 with one of these or maybe even two in parallel if necessary.

im being a little cautious cause i dont want to start a fire. id be willing to test it myself but i dont want it to burn the fuck down after like 2 weeks when im away from home and my mom is just using it herself.

space heater is rated at 500W/1000W (low/high) 120V AC 60Hz. also this copper strip/coil thing was rattling inside and came out when i opened it up. looks like just some random piece of copper it doesnt look functional at all but it was just sitting lose in the base of the space heater. will link pic in a second
>>
File: IMG_0045.jpg (1.12 MB, 2592x1936)
1.12 MB
1.12 MB JPG
>>1307145
pic related
>>
>>1307145
>1N4003 rectifier diodes that are rated for 200V 30A
Make that 1A. And you generally shouldn't parallel diodes.
>>
>>1307147
make that 1A? what do you mean?
>>
>>1307149
oh wait nevermind the forward current is 1A but the surge current is 30A. so i guess that won't work then
>>
File: 1512311047573.png (35 KB, 833x332)
35 KB
35 KB PNG
>>1307145
no. replace them with an actual 15A 600V diode. 240V non-repetitive reverse blocking voltage is way too low for a mains application and forward current is woefully insufficient.
>copper strip/coil
that's quite likely part of the thermostat. that heater sounds pretty fucked, m8. better replace the whole thing
>>
>>1307147
if i have a schottky diode with a forward current of 30A but only 45V peak reverse voltage, could i somehow use these two diodes to fit the specification?
>>
>>1307155
please don't touch anything mains related
>>
>>1307156
i havent yet just trying to learn bruh u wont be the one who dies anyways
>>
>>1307145
Read this application note on paralleling diodes so you don't burn down your mom's house to save 30 bucks on a new heater.
http://www.st.com/content/ccc/resource/technical/document/application_note/a4/ef/bc/7d/78/89/49/f1/DM00098381.pdf/files/DM00098381.pdf/jcr:content/translations/en.DM00098381.pdf
If you stop after reading the first paragraph and think "yeah it says right there you can parallel diodes no biggie" then you're a retard. If you kept reading and understand the danger, then bless you.
Diodes don't just automatically share current evenly because of manufacturing variations and especially because their forward voltage has a negative temperature coefficient.
>>
File: teensy touch.webm (1.78 MB, 332x585)
1.78 MB
1.78 MB WEBM
For my first arduino project I am making a touch based midi controller for my friend who does music stuff for fun. I have an issue though that I cant really seem to find an answer for nor can I think of a solution.
I am using a Teensy LC and the touchRead function to get the value of the sensors I made. They are just a grid of copper that plugs into a touch sensing pin.
Now my issue lies in converting the touch data into binary buttons. My first thought was to set up a threshold where if the value is above that it counts as on. That doesn't work well because different sensors and pins have a different baseline and different value when touched. My next thought was that when it first turns on it would wait a second of no touching to calculate the value it is when touching then you tough it and it calculates when it is touched then sets the threshold on a per pin basis at around halfway between the two values. That didnt work either because if hitting the sensor two quickly the value doesnt have enough time to fall below the threshold before rising again (webm related shows this).

Any thoughts on how to turn teensy capacitance sensors into binary buttons?
>>
File: diy teensy touch help.png (501 KB, 592x489)
501 KB
501 KB PNG
>>1307181
Oh and this is what the current testing setup looks like. The final version will have more touch buttons, potentiometers, and an led.
>>
I have 20mVpp noise after an LDO, at minimum load. It looks periodic, but the load is constant. I sort of need the smoothest voltage.
What can i do to reduce it? Adding more caps did nothing.
>>
>>1307210
The first thing that came to mind was that your capacitors aren't what the regulator expects. See if there are any requirements for ESR.
The second thing was that the noise isn't real, but comes from scope probe grounding or something similar. Check if you see same noise in unexpected places.
If the same shit is visible at regulator's input, you can try to improve input filtering.
>>
>>1307210
>I sort of need the smoothest voltage.
Whatcha working on?
Try using a battery instead to see if the noise problem is coming from your power source.
>>
>>1307181
your slopes look pretty smooth. try looking at the difference between samples and applying them to a separate register per pin clamped to some small fraction of the "travel". pseudocode:
int16_t pressure[NUM_PINS];
int16_t sens = 5;
...
pressure[pin] = min(max(pressure[pin] + val[pin] - lastval[pin], -sens), sens);
if (pressure[pin] > sens / 2 && !isDown[pin]) {
isDown[pin] = 1; sendKeyDown(note[pin]);
} else if (pressure[pin] < 0 && isDown[pin]) {
isDown[pin] = 0; sendKeyUp(note[pin]);
}
also consider an IC more finely tuned for touch sensing like the cheap and available TTP22x
>>
Ignition coil anon reporting.

I went back to Frys and found a FET that was rated for 400 volts drain-source breakdown and had the same avalange energy rating.

Holy fucking shit. This is now terrifying.

The potentiometer I was using to play with snubber values got set on fire.

With the other FET, I was using receipt paper to see how the arc activated the heat sensitive ink in the paper. With the new FET, it just punches a hole through the paper and then sets it on fire.

My scope says that the spike on the primary is something like 420 volts (maxes out the display on 5V/div and 10x on the probe).
>>
>>1306848
>seems kind of weird that the arc voltage would stay so steady throughout the discharge
It wouldn't, probably. The voltage required to strike the arc is a lot higher than the voltage required to sustain it. And the resistance of the ionized air within the arc is much much lower than the resistance of air.
>parasitic capacitances
this is why cars have a capacitor across the switching points in the ignition circuit. to absorb the inductive kick and ring with the primary.
>>
File: 1511003590710.png (185 KB, 766x721)
185 KB
185 KB PNG
>>1307321
noice
>>
>>1307348
I put some caps between the spike side of the secondary and ground and now it's EVEN ROWDIER. The spike takes a tiny bit longer to ramp up but lasts way, way longer. Like 30% longer. If the body diode of the FET wasn't leaking the negative voltage to ground, then this would be completely crazy. Right now I'm losing half of my energy that way, I think.

I can't wait until I order a proper IGBT and get another 200 volts on the primary. I'm having way too much fun with this.
>>
File: burntup.jpg (209 KB, 681x1210)
209 KB
209 KB JPG
First time posting, but figured I'd give it a shot to ask:

Was given a drone by a friend who said he got it a few months ago but it never worked, so he said I could keep it, even if I made it work. So I opened it up to see if anything looked fucked, and I found this on the underside of the circuit board. I don't know much about these things other than very basic knowledge, but is it safe to assume this is the problem or at least part of it?

Nothing else stood out as odd. Like I said, I don't know much about circuitry.
>>
>>1307356
Yep that's proper fucked.

If you bridge the melted trace with either solder or some copper, it might work for a little while. Gotta find out why the trace melted though, which might be harder. Might just happen again.
>>
>>1307355
you seem to be getting high off the ozone, my dude. you should hook up that C across your FET, start with like 100pF and try to get that peak down to 300V or so. you should still get a nice long lasting spark due to the aforementioned ringing effect. I don't think your caps are rated for those tens of kilovolts you seem to be getting on the secondary
of course, you can vary your duty cycle aka dwell time. you should do this and observe the effect.
>>
>>1307373
My bad, I meant to say primary. They're the little ceramic disc caps. I don't know the value but it's not more than 200 pF. They're tiny. I've already cooked a couple of them. Melted the leads right off.

>ozone
Shit I didn't think about that. I'll smell it by the time it gets to be actually hazardous right?
>>
>>1307356
I'd scrub that garbage off to make sure that the trace itself is what's gone bad and not some now vaporised SMD capacitor. To fix the broken trace just solder some solid copper wire (~0.5-1mm diameter) over the section.

Something might have happened to cause the trace to have way too much current through it, but it's probably more likely that the trace itself was bad in the first place.

In any case, post a picture of the reverse side of the board so I can see if there's anything else that stands out.

>>1307378
Won't the NO2 be in higher quantities than the O3?
>>
>>1307380
It doesn't look like there were any components on that trace. A surface mound part would be a long shot as well based on the selection of the other components.

What likely happened was a stray object was pressed too close to that trace and shorted it to ground, maybe melting it and demolishing the trace.
>>
>>1307398
That sounds likely.
>>
>>1307378
Ozone smells very very sharp, can't miss it, you're not going to hurt yourself from ozone inhalation even if you played with your zapper all day long.
>>
Does anybody know how to move a selection of stuff to a different layer in kicad, I want to move a bunch of component labels to Cmts.User
>>
>>1307355
Just be careful you don't fly too close to the sun. Remember that if the secondary voltage gets too high, it could break down the windings' insulation.

>>1307373
>you seem to be getting high off the ozone
I love that phrase. high voltage is an addictive drug.
>>
I bought some of these 20mm piezoelectric transducers https://www.aliexpress.com/item//32789654110.html? , but I can't get them to work.
Pic related is what I tried to drive them with. What am I doing wrong guys? I can't find anything online on how to drive these.
plz help
>>
>>1307784
>1.7MHz
doubt it. did they work alright on channel 2?
>>
>>1307784
Piezos tend to have very sharp resonances, so that if your frequency is off a little bit from the true resonance frequency, you can't get much power in.
Also, your signal generator's 50 ohm output isn't going to push much power to the element.

It wouldn't be a surprise if the element required additional stuff to work properly, like resonators or whatever.
>>
>>1307784
> I can't get them to work

how would you even know? can you hear sound at 1.7Mhz? maybe divide by that 1000.
>>
>>1307927
He should be able to measure the current going through it with a shunt resistor and a scope. If he adjusts the frequency until he gets a peak in the current then he knows it's working. He could also use another transducer as a receiver and hook it up to a scope and see from there if he receives a signal at all. I assume he's already done one or the other.
>>
File: us-generator.jpg (50 KB, 500x750)
50 KB
50 KB JPG
>>1307784
you need the power
>>
>>1307989
>>1307807
Thanks, I think power is the issue, I will try an h-bridge.
>>
Can anyone explain how the voltage works in soldering irons? I’m looking at buying a 30watt weller soldering iron that plugs into a 110v wall socket. the iron doesn’t use all 110v right? How does it step down the voltage so it doesnt explode? I want to know if I can rig my soldering iron up to a 12v battery and still have it work? Or does it need 110v?
>>
>>1308130
Ohm's Law. Current is equal to voltage divided by resistance. I=V/R
Power (watts) is equal to voltage times current. P=VI
and following from that, power is also equal to voltage squared divided by resistance. P=V^2/R

A soldering iron is just a big resistor, it has a specific resistance which determines its rated power at rated voltage.
The power of your soldering iron actually depends on the square of the voltage. it'll drop off a lot with just a small drop in voltage.
so your 12 volt battery will not power a 120 volt iron.
>>
File: 2011532225564.gif (42 KB, 600x308)
42 KB
42 KB GIF
>>1308072
lots of volts
>>
>>1308148
I mean it gets more complicated if the soldering iron has electronic temperature control, some nicer soldering stations do use a low voltage DC to run the iron.
real cheap irons just plug into a wall outlet and run at a constant power.
Slightly better ones have a power control, but that's just the same thing as a dimmer switch and the iron still runs on the 120 volts.
anything with a temperature control might have an iron that uses 12-48 volts, but in that case it still won't just run on a battery if you hook 12v up to the plug, the electronics need power.

If you're going to spend like $50 on a soldering iron anyway you should just get a mini electronic one that's made to run on battery power. like the TS100, or the Sain Pro32 which is a clone. amazon has these for like $50-$70.
those things are rated to run on 12-24 volts.
>>
>>1308130
Your soldering iron will not be able to run on a 12V battery, it needs an ac voltage that is quite a bit higher, maybe you could mod a soldering station that steps down the voltage and then uses it for the soldering iron and skipping the step down process but I would recommend just buying a battery powered soldering iron from somewhere like hobbyking.
>>
>>1308148
I don’t need the iron to run at full power , I want to take a cheap fixed temp iron and modify to use a rotary dimmer to control the power to the iron . Will a buck converter work if I hook it up to 110v source and then I can modify the voltage out from that ?
>>
>>1308130
Here you go, 12V soldering iron https://hobbyking.com/en_us/soldering-iron-30watt-12v-3s-xt60-plug.html?___store=en_us
>>
>>1308160
imagine it had a temp sensor..
>>
>>1308159
>I want to take a cheap fixed temp iron and modify to use a rotary dimmer to control the power to the iron
If that's all you want to do, use a regular dimmer switch meant for lights. You can even buy a module that has the dimmer and plugs built in, that you can just plug your cheap simple iron into.
You don't need to go down to low voltages and you wouldn't want to. The power of the iron goes by the square of the voltage. So, half the voltage will give you a quarter of the power. A quarter of the voltage will give you a sixteenth of the power.
If you take your iron down from 120 to 12 volts, that's one tenth the voltage and you'll have one one-hundredth of the power. Your 130 watt iron becomes 1.3 watts. It'll barely even get warm at that.
You don't need a buck converter, it's needlessly complicated and expensive. An AC dimmer that uses a triac is very low cost, low component count (literally like 4 parts).
It doesn't vary the voltage, it just chops it up so that the iron gets a small piece of it. it still controls the average power.
>>
>>1308180
Any links to the dimmer + plug module you are talking about? I can only find dimmers . Basically I want to start off by reducing the voltage in the system to the point where the hottest the iron can run is around 250C and then I want to have a dimmer on top of all that that can modulate the power further so I can adjust in a range from 0C - 250C. I was thinking I could use a buck converter to drop the iron to about 1/2 it’s max power and from there use a dimmer to control how the remaining power in the system. Can I set up a circuit that takes power from my wall outlet and runs it through an 8$ buck converter so I can play around with sending different lower voltages to the iron? You say buck converter is needlessly complicated but I’m willing to get complicated if it gives me the precise control I need
>>
File: dimmer.jpg (3 KB, 251x201)
3 KB
3 KB JPG
>>1308188
>>
>>1308188
Remember that power control is not temperature control. You will never have precise control without monitoring the iron temperature.
A certain power might be enough to hold 250c when the iron is in air, but as soon as you start melting solder with it, the temperature drops.
>>
File: question fox.jpg (124 KB, 620x451)
124 KB
124 KB JPG
Reposting from QTDDTOT, figuring I might as well take it directly to /ohm/.

TL;DR: Is there an arduino-tier choice when it comes to FPGAs? What is generally reccomended?


I've been working on a project involving torque control of a 3-phase (PMSM/ basically a bldc) motor.
This involves measuring the current in each of the 3 phases going to the motor, keeping track of a 10 bit rotational encoder, doing a few calculations and finally producing PWM to control three bridges for power to the phases of the motor.
I made the prototype work using a teensy 3.6 (180 MHz and arduous compatible) as processor, but it seems to be barely cutting it.
I'm going to take the project next level by adding a second motor with torque control as well, and then I doubt it'll be enough with the teensy.
The trouble is also that the currents should (ideally) be measured simultaneously and at just the right moment, relative to the PWM signals.
I know I can save a lot of time by adjusting the adc as well, but I'm curious about trying an FPGA anyway, not to mention the advantages of synchronous processing.

So... any recommendations on a "beginner-friendly" FPGA that can handle 6 analog inputs, 2 interrupts and a few digital general I/Os?
>>
>>1308249
FPGAs aren't sequential instruction processors, my dude... unless you design and instantiate one in HDL, e.g. picoBlaze. also they don't usually have onboard ADCs. you would need to supply your own.
anything cheap is good to get started. there's an Altera Cyclone II dev board that comes with a clone USB Blaster JTAG pod for about $20 on aliexpress. download Quartus Prime from altera .com and see if you can wrap your head around Verilog (similar to Pascal) or VHDL (similar to Ada). almost certainly an FPGA is overkill for this.
now, as to your application, you should consider throwing some hardware at it. try external sample-and-hold amps such as LF398 to satisfy the timing requirements of the current sampling.
more than likely, you have some fat in your motor control code that could be trimmed off, by precalculating some tables, using time-optimized algorithms e.g. the fast inverse square root at the expense of a little accuracy, going around the lolduino libraries and hitting hardware directly, and/or working in fixed point where feasible.
if it turns out you actually do need hardware assist, you should check out the line of STM32 MCUs, some of which have rotary encoder support on at least one timer/counter, multiple pretty fast ADCs, three-phase PWM, and plenty of CPU. they're cheap and pretty easy to use, especially with their STM32Cube software that writes the setup code for you and you need only fill in the blanks between comments. some higher-end devices in this range also have hardware floating point instruction support. some of the devices are available on dev boards.
>>
>>1303811
I want to change the cruise control on my car using my memeberry pi.
The cruise control buttons work as follows: there are two wires, pressing a button completes the circuit with a specific resistance, the computer reads the resistance in the wire and knows what button was pressed.
To do this with the pi, I was planning on using gpio to actuate a relay that completes the circut with the wires. I was going to use an ohmmeter to measure the resistances in the wires when different buttons are pressed and the add a similar resistor to my circuit with the relay. Would this work? I have never done anything with electronics before.
I don't actually have the necessary relays right now. I have some transistors that I was thinking I might be able to use, but I watched a youtube video on transistors and got confused. They add resistance to the circuit and I have no idea which way the electricity is supposed to flow in the wires to the buttons. Then I would also need to calculate something for an addition resistor so the raspberry pi doesn't get fucked up or something. I don't understand all this so I was just going to try using a relay. Would my relay idea work?
>>
>>1308317
Typically buttons don't short a resistor or add a resistor in series or anything like that, usually they just close an open circuit and the computer does the rest (debouncing and such). But often buttons will have a pullup/pulldown resistor on one end such that they are at some voltage before the button is pressed. Like pic related. Since a relay does the same thing as a button (open circuit when deactivated and closed circuit when activated) there should be no problem with replacing buttons with relays, or otherwise putting the relays in parallel with the buttons (assuming the buttons are normally open).

But if you think the Pi will be spamming those relays at more than a few seconds between each actuation I'd use MOSFETs instead, which might take some more design considerations.
>>
File: pullup resistor.jpg (10 KB, 400x293)
10 KB
10 KB JPG
>>1308331
Nyaa
>>
>>1308264
I will look into that, thanks for now anon.
>>
File: 1506633055575.png (1.1 MB, 754x692)
1.1 MB
1.1 MB PNG
>>1303811
I just came here to ask that how brainlet one has to be to have electronics as their hobby?
Assuming you have a stable job you could spend the free time improving your health by exercising, or enjoying the outdoors or hanging out with friends.
Instead of that you fiddle with resistors at home, sad.
>>
>>1308337
>buyfagging this hard
How's that unserviceable computer and radio and television doing for you? Oh you had to buy a completely new one when the power board broke? I just replaced a couple of capacitors and saved myself a few hundred.
>>
>>1308337
you will never enjoy the joy of using one of the most powerful and useful forms of art the modern world has to offer, using our current tools to best cater to your comfort, amusement, or simply to see how you can create, destroy or reshape what's around you.

But hey, i'm studying EE's, so i'm biased as hell. Everyone has their own interests. a chemist could probably copypaste what I said and believe it anyways.

Have fun, the world is a big one.

Plus, exercising is mandatory for everyone, regardless of hobby, to have a healthy mind
>>
>>1308337
>I just came here to ask that how brainlet one has to be to have gaming as their hobby?
>Assuming you have a stable job you could spend the free time improving your health by exercising, or enjoying the outdoors or hanging out with friends.
>Instead of that you fiddle with gamepads at home, sad.
Post this on /v/ please.
>>
File: Screenshot_1.png (160 KB, 1919x1079)
160 KB
160 KB PNG
What is the standard way of adding the pin numbers to a component in eagle?

I'm making a voltage regulator from an existing 78XX device, and i'm not sure where and how to put the pin numbers. The package has VIN,GND,VOUT. The schematic has no names on the pins, just names put as text (I asume because the box is so small using the pin name makes things awkward to read), and the "visible" tab is simply off.

When I put the device as is, it shows everything just as it looks like in pic related.
>>
>>1308331
I've seen it on things like power door lock controls. in fact, the variable resistance keypad is not an uncommon arrangement in devices made cheaply by the tens or hundreds of thousands that have comparatively long wire runs. some MCUs/SoCs have a low-resolution ADC intended precisely for this purpose.

>>1308317
relays should be fine, the only real concern being possible contact closure initiated by the physical shock of a pothole or whatever. don't worry too much about "spamming" relays, they should be fine with a couple of "presses" per second or more.
in the event the aforementioned bounce becomes a problem, and the resistances in the buttons are tied to ground, you could measure the resistance when particular buttons are pressed, and then short the same resistance values to ground with a MOSFET. be sure the cruise control signal line is disconnected from everything else when measuring or you might inadvertently include a pull-up resistance and trigger the wrong button.
>>
>>1308364
I think in that case the variable resistance thing is used because that's simply how the cheap resistive buttons work, not because it's in any way a desirable way to actuate a button. Though perhaps it removes the need for debouncing. In any case, I doubt standard cruise control buttons would be of the resistance sort.
>>
hey guys, i recently took a cheapo meter apart and found the buzzer responsible for the continuity check sound. i think its piezoelectric? it has two wire terminals. i tried connecting them across a battery, which results in a short click. it will not click again until i reverse the polarity of the voltage applied (some kind of springlike hysteresis with the metal inside?). what is the proper way to drive these? an AC voltage signal? thanks
>>
>>1308416


If I had to guess it was driven by PWM before, DC on and off at desired frequency. AC will do the trick
>>
>>1308416
Needs AC, the cheapo meter will just be driving a square wave by turning an output on and off.
>>
>>1308405
>just how they work
Ford vehicles, for example, use exactly this scheme of a button-dependent resistance to send cruise commands to the ECU which expects the voltage to fall into a given reference range for each command.
in the power door lock case, Chrysler uses this extensively. http://www.carwirefire. com/car_security_door_lock_configuration_single_wire.html
why? it saves millions of meters of wire, slip rings on the steering wheel, pins on the ECU. why wouldn't they?
get out into the world more, kid
>>
What is the best resource available to learn autocad from scratch? I want to use it to make small designs to 3d print to support and mount electronic projects
>>
>>1308457
>>>/3/
>>
>>1308456
>slip rings on the steering wheel
Do cars' buttons use slip rings on a steering wheel? Since it only turns a limited amount, flexible enough wires would probably work fine, perhaps with a load-bearing plastic insulation to prevent the copper itself from work-hardening.

I see now that you mean you can put multiple button signals along the same wire, instead of just having a single button that isn't closed or open but set up such that it has two different resistances depending on whether on or off.
>>
>>1308532
Steering wheel probably clock spring arrangement. I have seen wireless buttons for radio that needs a battery in the wheel
But horn has to be wired
>>
Didnt want to open a thread for this so asking here. Does this cap look bad?
Its from a tv psu which doesnt even light up the stand by led let alone respond to the power button.
>>
>>1308564
yes
>>
Hey guys, I am working on usb project, and I am looking for a "housing" similar to pic related, Mainly a housing that pre made to fit usb-A male and usb-A female, is there a proper word I can look up on ebay, and buy them?

thanks!
>>
>>1308629

uhm... what? You realize this keeps your interface node the same right?

Anyways I assume you want some sort of type converter.

https://www.amazon.com/AUKEY-Adapter-SuperSpeed-MacBook-Samsung/dp/B00W98IJ0O/ref=pd_lpo_vtph_147_lp_img_2?_encoding=UTF8&psc=1&refRID=17K3W6B2EZ737S063VAW
>>
File: usbenclosure.jpg (26 KB, 960x520)
26 KB
26 KB JPG
>>1308635
no no, I just want the enclosure, I actually found some results Pic related is closer to what I want.

I am looking for enclosure that can house usb-a male and female, no adapter.
>>
>>1308635
Samefag here, I found what I am looking for here

https://www.aliexpress.com/store/product/4-pcs-53-24-14mm-smart-small-usb-plastic-enclosure-for-electronics-plastic-case-diy-usb/1006252_32764299612.html?spm=2114.12010612.0.0.2b20a4cbf1tj9z
>>
File: 757128427_497.jpg (13 KB, 500x332)
13 KB
13 KB JPG
>>1308629
>>1308629
Like this?
Logitech wireless stuff came with them to extend the receiver past the combuter case.
>>
>>1308643
Take a look at my post here >>1308641

thanks for helping though, I found what I want, now, to see if this user on aliexpress is trustworthy.
>>
File: 1494128104356.png (12 KB, 761x673)
12 KB
12 KB PNG
>>1308532
548 is probably right, clock spring, which is still a couple of meters of polyimide. still, volume manufacturers do squeeze the hell out of BoM costs within the bounds of reliability and they will definitely prefer one wire to 5.
(resistor values brought to you by the JACAL symbolic math system)
>>
Hey guys I'm working on mini project and was wondering what I can use to add Wi-Fi to my project, and be able to do file transfer off if the eeprom chip. I was thinking famous 8266 chip, but is that chip arsonists only?
>>
>>1308854
8266 is bretty gud.
it's cheap, easy to get started programming it (in C, lua or python), there are arduino libraries for it, and it's popular enough that most questions you'd have probably are answered already.
>>
>>1308869
Well I'm gonna use a pic 12f1822 sonic.
>>
>>1308854
the ESP8266 does consume an embarrassing amount of power, yes, but it won't set the building on fire
recommend you have a look at e.g. NodeMCU, so named because of certain resemblance to Node.js including among others the HTTP API, and reconsider whether there's even a need for a PIC in your design
also disable your autocorrect pls
>>
>>1308945
I may not even use it since I'm looking for smaller solution, any sites you recommend I can filter Wi-Fi chips, and see what my options are?
>>
what
>>
>>1308963
It's a simple question
>>
>>1308963
Nvm, transceiver is the word I was looking for
>>
>>1308954
hm, I just searched for "qfn wifi ic" and came up with a few candidates, most weak. I suppose you could try looking for WLCSP packages instead or get some ideas for better search terms from those results
fwiw, most of the peripheral solutions expect an SDIO interface, which may be more than that 12F1822's flash and GPIO can fairly handle. others are full MCUs integrated with wi-fi hw
if a 5mmx5mm QFN32's too big for you, I don't know what to tell you, because you'll also need space to handle the RF signal, which has some critical geometry
>>
>>1308249
>I made the prototype work using a teensy 3.6 (180 MHz and arduous compatible) as processor, but it seems to be barely cutting it.
>I'm going to take the project next level by adding a second motor with torque control as well, and then I doubt it'll be enough with the teensy.
> So... any recommendations on a "beginner-friendly" FPGA that can handle 6 analog inputs, 2 interrupts and a few digital general I/Os?

You need a DSP! FPGA's aren't well suited for the task you're up to.
>>
>>1308249
As the other anon said before, your best bet is a motor control oriented DSP or DSC.
I'm currently using a F28335 to control a BLDC motor. Shit's good, but TI documentation can be a pita sometimes.
>>
File: newvcrbanner.jpg (107 KB, 1600x1200)
107 KB
107 KB JPG
>>1303811
Any parts worth scavenging (motors/etc) from these?
>>
>>1309206
Scavenge everything, including the smallest of capacitors and resistors.
>>
Hey /ohm/ I'm trying to repair the potentiometer for the sticks on my gamecube controller, but I don't have soldering braid or desoldering tools to remove the solder holding the potentiometer in place. You guys and gals got any tips or homebrew solutions for removing solder?
>>
>>1309307
Kitchen sponge, not even kidding.
>>
>>1309346
Would you believe me if I told I didnt have any?
>>
>>1308869
How can i control it with a pic? Almost every guide on them uses either arduino or some obscure programming language
>>
>>1309431
I'm another anon, and from what i've seen so far I'm fairly certain I do need a pic to put it all together.
>>
>>1309434
So far i've been able to control it with my computer using an USB-UART, serial communication, and flashing it back into the basic AT-commands, but I need to talk to it with a pic to take the next step.

And yes, i'm sorry, I wont use three messages to try and speak
>>
>>1309431
if you can't find the USART on the PIC, why not just drop the PIC and use the ESP8266 for whatever you were going to use the PIC for?
here, go blink an LED https://www.cnx-software.com/2016/10/07/how-to-write-esp8266-firmware-from-scratch-using-esp-bare-metal-sdk-and-c-language/
>>
>>1306542
Multiple reasons.

The value doesn't matter as much as the restive ratio.
These are laser trimmed to have the right ratio to a ridiculous spec.
They share a single package will means they share a similar thermal response with a low thermal gradient between the elements.
>>
>>1308564
In general, Electrolytic caps with any bulge should be replaced, even if they aren't "bad."
>>
>>1303811
Does anyone know anything about recapping old CRTs? How do I discharge caps?
>>
>>1303811
This might be a dumb question but is it bad to use rechargable batteries if I'm fucking about learning electronics and potentially creating shorts etc?
I'm pretty sure I read somewhere that they're a lot more dangerous if shorted due to lower internal resistance or something.
Obviously I'd rather use rechargables and save money if possible.
>>
>>1309547
Use a power rated resistor in parallel, and measure the voltage with your meter to make sure it discharges until it's safe (less than 20V).

If you don't have a suitable resistor you can literally "crowbar" it with a a tool. Just make sure the tool is insulated and make sure it's something you don't care to much about if it welds to the terminal or otherwise gets pitted.

Beyond that if you don't know how to discharge those high voltage caps you should not be fucking around inside of a CRT power supply. The voltages are lethal.
>>
>>1309568
When you replace the caps be sure to install proper bleeder resistors.
>>
>>1309565
LiPo's are dangerous in short condition. This is why vapers who brag about their "low ohm mods" end up blowing their fucking hands off.

Buy a cheap DC "lab" supply on ebay or amazon. You can buy supplies that will exceed your requirements for sub $30 and they are useful as fuck. You can use them up to 30V and a few amps.

It's safer, it's more versatile, and you can limit the current for those times you will inevitably short somethign out while learning.
>>
>>1309571
Thanks anon, I'll stick with disposables then.
A bench supply sounds nice but I think I'd rather just spend more and get a good quality one later on than getting some chinkshit now.
I did see pic related when browsing Ali earlier though, which is a breadboard "power supply" for less than $1 which has two rails of either 5 or 3v.
I don't know how good the quality is but if I bought some 9v batteries I could then basically get 3 different voltages from the same cell.
>>
File: breadboard_psu_MED.jpg (94 KB, 500x375)
94 KB
94 KB JPG
>>1309582
Forgot pic
>>
>>1309582

The thing about chink shit is that, while lots fo ti may be shit, the stuff that is ridiculously common like lab supplies are typically stolen designs. You can basically buy an American designed lab supply for nothing, you just have some engrish name on the supply and the plastic is a different composition than the name brand.

The other option could be building your own. Should be enough info on the net or even kits to buy.

Those breadboard ones are OK, but I'm not sure they have any real protection for shorts.

Between the old equipment on ebay, and the chinkshit on Amazon, it makes electronics much cheaper/easier to get into these days.

Just stay away from rechargable batteries, and chinkchargers.
>>
File: 1508336450884.jpg (122 KB, 640x640)
122 KB
122 KB JPG
>>1309582
Power from the bench supply or equivalent when there's a chance of shorting. When there's no chance of shorting, then you may use rechargeable batteries.
There are chinkshit modules such as Pic related that include basic battery protection (overvoltage, undervoltage, overcurrent) and a micro-USB charging circuit for an Li+, all in one. I wouldn't depend on them against fat fingers though. Be careful that you monitor your first charge with one of these in a safe location, there are sometimes dud chips. To learn what is a safe location, search "lithium ion short" on youtube.
>>
2 basic questions

easiest yet decent way to get a blinking led

AND since diodes have reverse leakage could i use say 3 in series to prevent ANY reverse voltage/current. by any i mean less than a nano. like open-circuit basically. what is the result of 3 diodes in series.

thanks guys. learning so be nice please.
>>
File: 1N4148.png (66 KB, 1072x504)
66 KB
66 KB PNG
>>1309613
>3 diodes in series
see for yourself
>>
>>1309635
im learning so that pic is confusing
>>
>>1309613
>blinking led
Buy a blinking LED, use a relay and make a racket, make a shitty reverse-BJT relaxation oscillator, make a 2-transistor astable multivibrator, use a 555 timer, use a compatator/opamp oscillator, use a fucking arduino in order of decreasing simplicity.

>diodes
A diode follows an equation to relate current to voltage; the Shockley diode equation. Essentially, I = A*(exp(B*V)-1), where A is really large and B is really small. Essentially, this means that a small change in the voltage across the diode will result in a large change in the current. The most extreme example would be 0.5V to 1V, where at 0.5V you would have barely a few mA or even µA, and at 1V the diode would try to pass literally hundreds of amps, burning itself out in the process.

Since voltages add in series and currents are equal in series, you simply add the voltages of each diode. Assuming all diodes are equal, you simply multiply this voltage by 3, being the voltage which corresponds to the current you're using. When driving an ohmic load (V/I = constant) with a diode in series, you don't really have to worry about the current through the diode because the resistor will limit the current that can flow through it, but if you want to power 3 LEDs (or laser diodes, etc.) you'll need to watch out how much current you drive them with, because if you simply hook them up to a DC power supply or battery the voltage could easily be 0.5V higher or lower than it should be, meaning you could either blow the LEDs or not light them much at all. For simple or low power applications a single resistor will limit current for you, but proper setups use constant-current linear regulators or even switched-mode regulators.

I didn't actually read the diode question. The real answer to your question would be: look it up in the damn datasheet. Reverse current doesn't change much off voltage (unless you're avalanching) so 3 diodes in reverse won't be much better than just one.
>>
>>1309662
thanks anon. ill prob try 555 timer first if its decent. arduino i can use already but that is overkill so im trying to avoid micros. basic components are what im trying to get working. as for diodes ill stick to 1. thanks again! keeping the DIY alive.
>>
Anyone know a good starting point for picking the right hall sensors for analog inputs on a micro?
Planning on building some flight sim gear with a Teensy and trying to narrow down my choices...
>>
>>1309677
C'mon, you know you want to...
>>
>>1309717
absolutely discharging
>>
anybody know a good resource (preferably books, but that's just my bias) to learn about microcontrollers, specifically PICs and embedded C programming for it? From scratch; interrupts, timers, registers, Communiations (how to set up I2C), etc...I've seen some on the internet but nothing really convincing
>>
>>1309907

- find a pirate bay proxy (like fastbay.net)
- type ''pic programming''
- download
- ????
- profit
>>
File: bump limit reached.png (15 KB, 640x480)
15 KB
15 KB PNG
>>1303811

thread has reached bump limit.
where's the fag that makes new ones?
serve me now, fag.
>>
new thread
>>1309922
>>
>>1309500
>>1308566
Yeah thanks. Hopefully its the cause for the problem. But holy shit why is mouser shipping so expensive?
>>
>>1309915
chill, faggot, this is slow board. it'll stay on page 5 for several hours to a day.
anyway
>>1309922
>>
does anyone know what wattage of current sensing resistors do laptop batteries use? i burnt one 0.05 ohm resistor, and besides resistance i only know its packaging, as its 2512 smd resitor. on aliexpress there are 1, 2 and 3w resistors but the 2 and 3w ones are too expensive
also, whats the difference between a current sense resistor and a regular resistor with small resistance value?




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

[Enable Mobile View / Use Mobile Site]

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