/ohm/ - electronics general

[Anonymous No.1242891]

bump limit reached on old thread >>1239434

https://www.wiki.printf.pl/index.php?title=Pasta (fresh)

>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, 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?
Circuit Wizard
ExpressPCB
EAGLE
KiCad

>Q1. Lithium batteries
Will explode if you abuse them. Read and understand all relevant datasheets and be prepared for catastrophe. See
batteryuniversity.com/learn/
>Q2. My circuit doesn't work. Halp?
Check wiring, soldering, part pinouts, and board artwork if applicable, then post schematic.

[Anonymous No.1242898]

>>1242891
>proudly bipolar
yess

[Anonymous No.1242903]

>>1242898
Mostly because I was out of solder fodder and bump limit was just too close not to.

[Anonymous No.1242909]

>>1242903
Eh, it was only page 6.

[Anonymous No.1242923]

>>1242891
Can broken solar panels still be used? They still give a voltage reading under sunlight.

[Anonymous No.1242924]

>>1242923
Sure, if you care enough to solder them back into a circuit.

[Anonymous No.1242926]

>tfw the constant current driver can't get enough voltage to my LED to run it at maximum current
>tfw I'll have to let it heat up enough for the thermal runaway to get it up to max current
>tfw the tiny leads on the LED broke off
It never ends.

[Anonymous No.1242936]

>>1242924
Other than reduced current, what's the catch with doing this?

[Anonymous No.1242938]

>>1242936
Pretty sure you can't have any broken segments in series because they'll act as reverse diodes and won't conduct. If you manage to get partial cells in series with complete cells, I can't imagine putting the same current through the lot will work, so chances are you'll be generating excess heat somewhere. To be safe I'd saw off any segments of the panels that aren't complete, provided it's easy enough to isolate the individual segments.

[Anonymous No.1242947]

>>1242938
So I can't use a broken half fragment in series with 9 unbroken panels?

[Anonymous No.1242953]

>>1242947
Don't take my word for it, but if I think of it like a bunch of LEDs in series, one of which is only suitable for half the current, that's where it gets me.

[Anonymous No.1242969]

Are either of these worth getting? Is there any functional difference?

[Anonymous No.1242972]

just found out 60/40 solder is bad for you is this a good fume extractor?
https://www.amazon.co.uk/gp/product/B01HDHEK8U/

[Anonymous No.1242978]

>>1242972
Sure, as would any of the cheaper ones without the arm. There's not much to a fume extractor, really. A fan and a filter. Buddy of mine made his own out of a spare muffin fan and a cardboard box with an outside vent. He also put his ashtray in it, kek.

>>1242969
What are you going to use it for?

[Anonymous No.1242980]

>>1242978
ok thanks, how often do you change filters on them? also went for desk mount since I don't have a lot of desk space

[Anonymous No.1242983]

>>1242965
See if you can get inside one and take some pics.

>>1242980
No idea, I just blow the fumes out of the way.

[Anonymous No.1242987]

>>1242953
Makes sense, I left all the broken ones aside for the moment.

I think I fucked up though. The Al foil underneath is connecting all the negatives together and there's blue wire on top connecting the positive in series to increase the voltage.

The voltage is staying at 0.5v per panel though and not adding up to 5v. Is it possible that this isn't series? Can't check the current because it's night time here.

Is there an extra step for series I'm missing? This is my second attempt at an electronics project and it's not working as well as I'd hoped...

[Anonymous No.1242991]

>>1242972
>Solder composition may vary but is commonly 60-63% tin and 37-40% lead.
>Because of the relatively low temperatures in electronic soldering, fumes from these metal constituents themselves are not normally a concern.
Just something I found online from a fairly reputable source, so chances are it isn't the Pb-Sn solder specifically that's going to mess with your lungs, rather the flux smoke. I just open a window and go easy on the flux, but if you're doing a lot at once then I can see why you'd want a fan.

>>1242978
General low voltage hobby stuff, nothing fancy. A $10 multimeter is doing fine for me (except for reliability issues) so I really only need something to tell me when I'm getting a frequency and what amplitude it is.

>>1242987
Connecting all the positives together and all the negatives together is putting the panels in parallel, not series; the voltage won't add but the current will.

[Anonymous No.1242992]

>>1242991
thanks

[Anonymous No.1242996]

>>1242991
Oh my God thank you so much.

Can't believe I didn't see it myself. Thanks Anon.

[Anonymous No.1242999]

>>1242996
And I broke 6 out of ten panels trying to get them off.
Really glad I bought a 50 pack.

[Anonymous No.1243124]

>>1242969
those DIY kits are not bad
the second one is likely the same exact thing as the first one, but with a case
because they also sell the case separately

[Anonymous No.1243137]

I have a voltage between -1V and 1V. Now I want it to change to 1V as soon as it's on 0V.
I hope this picture can explain it better than I do.

[Anonymous No.1243139]

>>1243137
>I hope this picture can explain it better than I do.

In general, the dependent variable should be on the y axis, opposite from what you have drawn.

does Vin always go from -1 to +1? Do you want Vout to be -1 or what before Vin hits 0?

after Vin passes 0, does Vout stay at +1 forever, or only while Vin is "on 0" as you say, or while Vin is greater than 0.

Your question is very hard to understand. Your drawing is very hard to understand.

Maybe I'm being slow. Sometimes that happens to me.

[Anonymous No.1243140]

>>1243124
>likely the same exact thing as the first one
No it isn't, more like a follow-up version. There's extensive documentation at the manufacturer's site at http://www.jyetech.com/Products/LcdScope/e150.php
I think it's one of more interesting 200kHz toy scopes, especially because of the rotary encoder that spares you the silly button pushing to change parameters. It has no internal battery (requires 9V 120mA) but you can run it from a single LiPo (or any 5V source) plus a small step-up converter and use it 'floating' like a multimeter.

[Anonymous No.1243141]

>>1243139
Dammit, I messed up.
Yeah, Vin should be the horizontal axis.
Vin ranges from -1 to +1, depending on the variable resistor that I'm using. So it can go back and forth at any time.
And no, Vout should be able to change back to -1 (or anything below 1V in general), in fact:
while Vin >= 0V then Vout = 1V

Thanks for your fast response!

[Anonymous No.1243142]

>>1243141

what you are describing is a comparator. go read about them and then ask more questions here if you like.

In general it's a bit easier to work with voltages that are all above zero, but there's nothing wrong with what you are doing. You might need two power supplies instead of one, is the main thing.

[Anonymous No.1243172]

I want to make something like The Mosquito https://en.wikipedia.org/wiki/The_Mosquito
without spending a gorillion dollars on the name brand shit. I was thinking of using this piezo buzzer
http://www.mouser.com/ProductDetail/DB-Unlimited/P13165-1/?qs=sGAEpiMZZMuTkJYgZlQcSYniBjjB6KcIQKb6hpFworELMzZt%2fleFGQ%3d%3d
but then I realized I could get a tweeter instead. Anyone got any thoughts on which will give more volume for my buck?

[Anonymous No.1243206]

>>1242978
>He also put his ashtray in it, kek.
smokes
worries about solder fumes

[Anonymous No.1243270]

>>1243172
I'm really not interested in your session ID but this is not a 'buzzer', it is a 16kHz bender that will not do much unless built in a horn and powered by a suitable generator. Which one are you planning to use? Can you post a circuit diagram?

[Anonymous No.1243278]

Do I really need a logic analyzer, or can my scope handle it?

[Anonymous No.1243283]

>>1243278
If you have to ask, then most likely not.

[Anonymous No.1243293]

What do you think about the nixie clock I'm designing? What would you change? What features would you add?
>DS3231 for ultra accurate timekeeping
>Atmega328 or Attiny2313 based, 20Mhz
>555 based boost converter (any better option?)
>Multiplexing with shift registers and transistor arrays
>Tilt sensor as a way to set times and alarms by tilting the clock or:
>VL53L0X range finder for inputting data by waving your hand

[Anonymous No.1243312]

>>1243293
>What would you change?
probably doesn't need to be 20mhz
are shift registers necessary vs. larger uc mux, what is cost hit?
how will tilt sensor work, good idea to be shaking it around aren't they fragile. lol fucing range finder so you pick a time how do you confirm? holding your hand still or something?
how many alarms, how do you tell if they are on/off?
what is alarm tone? music? radio?
get rid of the nixies lol
put in a radio time receiver.

[Anonymous No.1243313]

>>1243293
ESP8266/32 for time synchronization and inputting alarms from phone or computer.

[Anonymous No.1243322]

>>1243270
>Which one are you planning to use?
What do you mean? For the bender, the one I linked. If I chose a tweeter it'd be whatever's cheap, like Power Acoustik NB-2.

I haven't drawn up a schematic yet because I want to know which option is louder. I have enough design knowledge to be able to get either option to its maximum power rating.

[Anonymous No.1243323]

>>1243206
He had family. Venting whatever fumes outside is nothing short of considerate.

>>1243293
>nixie
Tired meme. The new hotness is lasers with galvos, kek
>555 based boost converter (any better option?)
How about the PWM built right into your micro?
>Multiplexing with shift registers and transistor arrays
There are single-chip solutions for this.
>input
I'd use capacitive touch switches, myself.

>>1243312
>are shift registers necessary vs. larger uc mux, what is cost hit?
Drivers would be necessary. Not that someone building le nixie clock would be too cost-conscious, I imagine.

[Anonymous No.1243324]

>>1242983
ok man. I'll will be a little busy to do it until next week but these threads are slow etc

[Anonymous No.1243334]

>>1243312
>get rid of the nixies lol
Cold cathode will never die.

I'd make a single knob with a button in its middle do the adjustments, with a full-rotation rotary encoder, preferably 5 bits or more. By holding the button for a couple of seconds you'd enter adjustment mode, and rotating the knob would adjust the current digit(s). Pressing the button again would go to the next digit(s). I think you could make it quite attractive, and it's got that nice physical detent. Radio time receiver is also a good idea.

[Anonymous No.1243347]

>>1243334
>push button
>rotary encoder jumps
fucking dreadful

[Anonymous No.1243377]

>>1243347
Better than anaemic touch buttons or tilt-proximity meme sensors. And it lets you adjust forwards and backwards, meaning you can twist 10 increments backwards instead go having to cycle through 50 button pushes. What do you suggest?

[Anonymous No.1243383]

>>1243377
Encoders with a built-in push switch are probably fine. If you want to be clever you could simply ignore the first pulse if there hasn't been one for some amount of time.

[Anonymous No.1243388]

>>1243383
But how would you switch from each set of digits if not with a button? You'd have to have 3 encoders, which isn't necessarily a bad thing.

[Anonymous No.1243397]

>>1243388
You push on the shaft to actuate it.

[Anonymous No.1243399]

>>1243397
Then why ignore the first pulse if there hasn't been one for some amount of time?

[Anonymous No.1243402]

>>1243399
I meant the first rotation pulse, in case the push also happens to coincide with a step as >>1243347 feared.

[Anonymous No.1243403]

>>1243322
>I want to know which option is louder
Connect both to your signal generator and listen.
What do you expect from a 13mm disc that bends itself back and forth at 16kHz in free air?

[Anonymous No.1243422]

>>1243293
A Nixie clock is supposed to be a visual object. I'm missing a design rationale, a 'personality', a raison d'être. Nothing is more dull than numbers in lockstep, perfect symbol of the one-track mind. Technical interna are unimpressive in a time where everyone has a device that synchronises itself to some atomic reference somewhere in space or on the globe while the iconic clock still has a face and still has hands.

[Anonymous No.1243520]

>>1243312
>probably doesn't need to be 20mhz
My experience tells me brightness is reduced is the multiplexing frequency is not high enough, I'll do some experiments though.
>how will tilt sensor work
You just tilt the clock a bit, depending on the angle the numbers will increase or decrease faster, allowing you to set time without buttons.
>lol fucing range finder so you pick a time how do you confirm? holding your hand still or something?
You can move your hand up and down the beam of the rangefinder, when the number is ok you get your hand out of the beam, the sudden change of distance will input that number.
>how many alarms, how do you tell if they are on/off?
I guess I could add memory for 10 alarms, and a eeprom chip to store those alarms even without power (maybe in the 2313 module). Thanks for the idea.
>how do you tell if they are on/off?
I coould use the neon bulb separators as indicators or add LEDs behind the nixies to tell if an alarm is on or not.
>what is alarm tone? music? radio?
My level of experience doesn't allows that. A piezo will be connected to the back of the case, different alarm tones and volumes could be selected.
>>1243313
I bought an ESP8266 for that purpose, but I still have a lot to learn before I can use it, maybe in the next versions.
>>1243323
>How about the PWM built right into your micro?
The micro already has to deal with controlling the shift registers.
>There are single-chip solutions for this.
I know, I have some of them, but they're more expensive. The only inconvenient is a bit more complicated layout.
>Not that someone building le nixie clock would be too cost-conscious
Lel, I bought the cheapest nixies at $3 each?
>>1243422
What other clock uses the fourth state of matter to display the time? A flamboyant design for the sake of it isn't better than a simple and efficient design.
>Word clocks
Everyone and his dog has one, they're visually appealing, but impractical.

Thanks for the input everyone.

[Anonymous No.1243521]

>>1243520
>The micro already has to deal with controlling the shift registers.
You need to read a datasheet, anon. Look under the "timer/counter" section. On almost every microcontrolller in existence, there is dedicated hardware for generating a PWM signal, often an ADC, and sometimes there is more than one.
>multiplexing
Wait, why, again? You've got shift registers. High-side switches at 180V are a pain in the ass. You sure you want to?

[Anonymous No.1243524]

>>1243521
>You need to read a datasheet, anon.
Yeah, I know, I should start using interrupts and timers, being used to arduino I don't give my brain much use.
>>1243521
>Wait, why, again? You've got shift registers. High-side switches at 180V are a pain in the ass. You sure you want to?
Why not? multiplexing would allow me to use 2 or 3 shift registers, compared to 6 or 7 if I don't. I would also reduce the pinout coming from the nixie tube board, the only downside would be higher use of processing power and maybe lower display brightness.
>High-side switches at 180V are a pain in the ass.
I think the ULN2003 is low side. Also the nixie tube kinda acts like an LED, the voltage drop and the transistor will only see a fraction of the voltage. Dave explained it in his nixie sub counter.
https://www.youtube.com/watch?v=7uogKucrPks

Dave didn't multiplex though, he used a bunch of expensive HV-shift registers instead of using a couple while multiplexing.

[Anonymous No.1243531]

>>1243524
>Also the nixie tube kinda acts like an LED, the voltage drop and the transistor will only see a fraction of the voltage.
Not necessarily when they're off. Transistors' Vce(bo) becomes very important.
>I think the ULN2003 is low side
Also, if you're multiplexing, that implies you're doing an AND function on the low side, where it gains you nothing in drivers and loses you lots in logic, or switching on both the high side and the low side, which requires separate high-voltage transistors and some extra arrangements to reflect the switching signal up to the high side.
>higher use of processing power
Somewhat, though you'd be fine with a 2MHz clock or so if you used interrupts well and built a PWM regulator into the AVR.
Don't try to run the multiplex or anything else timing-sensitive with delay() calls. It will just lead to sadness.
>and maybe lower display brightness
Definitely, in fact, 1/6 of what you might otherwise expect. That doesn't sound very legible in any sort of ambient lighting. You might want to get one running at 1/6 duty cycle to see what you're setting yourself up for.

[Anonymous No.1243540]

>>1243531
I'm working on a similar thing, and I just plan to use 10 mmbta42s for each tube, it's much simpler than any multiplexing, and cheap at like 5 cents each.

[Anonymous No.1243595]

>>1243422
>personality
A tube is a tube is a tube.
(Time, Date, Temperature)

[Anonymous No.1243599]

>>1243595
Dutch klok inside.

[Anonymous No.1243604]

>>1243599

ants dont read schematics.

[Anonymous No.1243635]

>>1243604
True, ants have a different visual apparatus akin to a spheroid sensor with no lens and a limited resolution (which is why they mostly rely on their ant-ennae), but in contrast to some of us they are quite able to recognise structures.

[Anonymous No.1243645]

hey fellow /g/'s
i want to look after a used analog oszilloscope as long as i cant get one.
What are some things i should search for or avoid?
i thought about >2 Channel and external trigger, do i need more?
what frequency range should i look for?

[Anonymous No.1243654]

I have a programming related problem.

When trying to setup fast PWM on Attiny 13A I'm fucking something up. I wanted to control cooler with it, but I'm testing it on a diode and the results are fucked up.

The diode is connected to the port with anode, then through resistor to ground. Obviously I've read the documentation and at least I think so is set up as I've should. Instead however the LED lights up regardless of whether OCR0A(duty cycle) is equal to 0 or 255 and the insensitivity doesn't change at all. Code in pic, if you want I can post register description too.

[Anonymous No.1243656]

>>1243654
Ignore zeroing that COM0A0, it was part of troubleshooting that didn't made it work.

[Anonymous No.1243670]

>>1243645
Better look _for_ a used scope as long as you _can_ get one
2CH + TRIG 20MHz 10/1 Probe. Any scope is better than no scope.

[Anonymous No.1243671]

http://www.doradolist.com/

May be interesting for some.

[Anonymous No.1243674]

>>1243654

Looks ok to me except FOC0A should not be set to 1, but that might not fix it.

[Anonymous No.1243682]

>>1243671
gems->electronics->1entry->*click*->amazon->aaargh

[Anonymous No.1243689]

>>1243674
You are right, the documentation says to force it to 0 and my brain fucked up.

Didn't fix the problem tho. Whatever, will try tomorrow, maybe I'll figure something out. One addition - for whatever reason, if I try to measure voltage between the output pin and input pin it goes straight to 0, which is weird because it should be something like 5*<duty cycle>.

[Anonymous No.1243699]

What is the correct way to install these things when there is no cabinet or shelf I can mount it to, seems like a bit of a safety risk in the bedroom where mains wires could get ripped out.

Could I get away with a regular AC adapter or do I need an LED driver? My strips are rated for 12v DC and have little resisters built in and I only need about 3 amps.

Are there any differences between constant voltage LED drivers and regular AC adopters?

[Anonymous No.1243701]

>>1243699
>adopters
adapters*

Has anyone got some of these running with a plain old AC adapter?

[Anonymous No.1243732]

>>1243699

the ones with current limiting resistors just need a 12V supply capable of supplying enough current. when you use high-power LEDs then resistors would be too hot and too wasteful, so you need a current-limited supply.

[Anonymous No.1243743]

>>1243699

as for mounting, just shove 'em under the sofa or beneath (or above) the furniture. it's what jesus would do. satan might opt for adhesive velcro.

[Anonymous No.1243753]

>>1243732
Thanks, so there won't be any ripple problems? I just read this and it seems constant voltage LED drivers are built to certain specs.

https://forum.digikey.com/t/constant-voltage-led-driver-vs-ac-dc-power-supply/390

I was looking at this one because it'll just plug into the wall, is fully enclosed and has several approvals so at least I know it won't go pop like a cheap Chinese adapter from amazon and the ripple is speced at 1.3% http://uk.rs-online.com/web/p/plug-in-power-supply/7316068/

I need 2.8 amps continuously but don't know if these sorts of adapters are meant for near max load all the time.

[Anonymous No.1243834]

>>1243753
The ripple in question has a frequency of 30..50 kHz. Do you expect to see that? You barely notice the 100/120 Hz flicker of a traditional fluorescent tube or a traditional light bulb.

I would prefer the 6.3A LED power supply because it will run on only about half its rated power and probably produce less heat than the 12V/3A adaptor.

Regarding installation, simply apply common sense. Don't obstruct air flow, let it run and touch it now and then.

[Anonymous No.1243842]

>>1243645
Dual 20MHz with external triggers were ubiquitous twenty years ago.

>>1243701
>safety risk in the bedroom where mains wires could get ripped out.
You wild anon, you.
>What is the correct way to install these things when there is no cabinet or shelf I can mount it to
Anchor screws work, if your wall supports them and you have your cables otherwise fastened to the wall. It looks like that supply has some sort of clamp holding the wires fast into the casing anyway.

>>1243753
The dropper resistors serve the function of the LED "driver". He's comparing against an old-style 60Hz transformer-rectifier-capacitor wall block, not a modern switching power supply.
>flicker
Non-issue with high-frequency switchers
>ripple
Ditto
>efficiency
Ditto for the average small installation
>surge protection
No more an issue than with any other power supply.
>and dimming
Do that externally
>I need 2.8 amps continuously but don't know if these sorts of adapters are meant for near max load all the time.
The 6.3A supply should be fine. Virtually all components last longer when derated.

[Anonymous No.1243853]

>>1243842
>The 6.3A supply should be fine. Virtually all components last longer when derated.
>>1243834
>I would prefer the 6.3A LED power supply because it will run on only about half its rated power and probably produce less heat than the 12V/3A adaptor.

Dam, I was trying to save a bit of cash by not buying a massively overrated adapter.

[Anonymous No.1243857]

>>1243853
There's a 40W adapter on RS that's a few quid cheaper. It's down the first page. The extra 13% should help.

[Anonymous No.1243870]

>>1243853
How much is the 'cheap' 6.3A adaptor and how much the RS stuff in relation to the LED strip?

[Anonymous No.1243873]

>>1243857
Thanks, I saw that but it appears to have much worse ripple. I have an radio right next to it so want to reduce ripple.

>>1243870
Light strip was about £15 a few years back, I'm looking to spend £20 max on a good 12v adapter for them. By good I mean safe and has passed certifications.

[Anonymous No.1243879]

>>1243873
>I have an radio right next to it so want to reduce ripple.
check your math, ±1.3% > 120mVp-p

[Anonymous No.1243882]

>>1243879
Can't find it now but the one he was referencing to was speced at 10% ripple voltage, which is too much imo.

I don't "see" ripple but can notice it in the form of strobing when I move around.

[Anonymous No.1243887]

>>1243882
Me too, but I'm fairly certain neither of us can see ripple in the few tens of kilohertz range. Anyway, here's the item I meant.
http://uk.rs-online.com/web/p/plug-in-power-supply/9032902/
Note also that the EMI filter is at the supply end of the cable rather than the load end, and also that LEDs themselves do not radiate EMI to disrupt radios like fluoros do.

[Anonymous No.1243895]

>>1243887
That's weird, they say 40w it can only do 3 amps. Hate it when the datasheets have errors.

Looks like its rebadged at rapid too https://www.rapidonline.com/pdf/559996_v1.pdf

[Anonymous No.1243929]

>>1243887
>Note also that the EMI filter is at the supply end of the cable rather than the load end
That's to block noise being emitted from the supply.
If the filter was on the load end then the wire would act like an antenna to radiate the noise.

[Anonymous No.1243936]

LED guy here, ok so it turns out I'm actually pulling closer to 2.33 amps so I guess the 36w adapter will be ok? I plan on putting a diode in series with the LED feed in order to drop the voltage and current slightly as I've read that these LED's are rated to within an inch of their life and can get dim with age. Maybe even come up with high and low arrangement with multiple diodes (PWM will be very bad for my radio signal).

Secondly this long ass test lead is dropping about .5v @ 2.33 amps so it will be limiting the current as these LED strips are in a very non linear curve at this stage, at 9v they have a nice gentle light to them whilst at 12v its like a flood light. The final cable will be about half the length of this one.

At 12v a regular diode in series seems to drop it to 1.76 amps whilst a schottky diode takes it to 1.833 amps. In this small room the difference in brightness isn't actually that noticeable so I guess once they reach a certain running voltage it just turns into more heat with very little light gain.

tl;dr will closer to 2.3-2.5 amps max be ok long term for the 3 amp adapter?

[Anonymous No.1243941]

>>1243936
Wow, just did some calculations with the help of my 30 milliohm current shunt and it turns out that a single diode saves several watts of power consumption with a not very noticeable difference in light level.

I guess these resistor ballasted strips are just not very efficient compared to the LED bulbs you can buy for a regular fixture.

[Anonymous No.1243945]

>>1243895
>Hate it when the datasheets have errors.
Especially careless, easy-math ones.

>>1243929
Yes, anon was concerned about the lighting being near radio apparatus.

>>1243936
>read that LEDs are rated
You can actually test this, you know. Each of those cut lines on the strip delineates a single circuit which is placed in parallel across the supply. Count the number of sections and divide by the amps. 20mA per section is the traditional maximum recommended LED current but nowhere near absolute maximum. If they are running in excess of that, some extra voltage drop might be useful in prolonging life.
>tl;dr will closer to 2.3-2.5 amps max be ok long term for the 3 amp adapter?
Sure. You could even PWM them at 2kHz at full brightness (LEDs usually have a somewhat greater pulse rating of at least 150% Ifmax)and probably not see any flicker or wavering.

>>1243941
Eyes adjust to the light they're provided. Also, don't forget that the diode is actually dissipating power.

[Anonymous No.1243947]

I need to bunch all of these wires together so I can power some stuff safely. Is there a better method than twisting them together and soldering? Or is that what /ohm/ would do?

[Anonymous No.1243949]

>>1243947
/ohm/ would use a properly sized wire. Did you calculate the current draw vs ampacity of the wire?

[Anonymous No.1243950]

>>1243947
I'd suggest a linesman splice for just two wires, but there's got to be a similar splice for connecting up to 8. You could always use a terminal block or a bus-bar with eye connectors if you want, or daisy chain those male+female spade connectors. Note that you can't simply put heat-shrink over a multi-wire splice, so it really depends on how much current capacity and insulation matters. What voltage/current and what for? Automotive?

[Anonymous No.1243954]

>>1243949
Those wires are coming out of an ATX power supply, so unfortunately I didn't get to choose their gague. And no I did not do any calculations, but I know this will be enough as this mod been done a lot before with far fewet wires.

>>1243950
It's to power a 3D printer, all of those wires are getting shoved into one of these RAMPS connectors.

[Anonymous No.1243957]

>>1243945
These is the led strip and I currently have 3.10 meters installed https://www.amazon.co.uk/gp/product/B008EN225Q/ref=oh_aui_detailpage_o05_s00?ie=UTF8&psc=1

Mean't to be 14.4w/meter but that is never the case IRL because of the resistance of the tape tracks and feed wires and their non linear voltage to current curve. They are just to light under a high sleeper bed because the ceiling light can't shine under here and its always dark, I did have it powered with an old bench PSU in the past but its a bit messy and is a linear regulator. Last thing I want is to come home and mess around plugging in some 30 year old PSU to light my room.

I've seen some of those wall PSU's have standby losses of under half a watt so I could simply install a switch on the 12v side and not worry about idle power being wasted.

The diode is only dissipating about 1.3w right now so its worth it for the saving of several watts, plus the light difference isn't that noticeable to me.

[Anonymous No.1243958]

>>1243947

that's what i would do, but i have an iron capable of 60W. if your wibbly wobbly iron can only do 30, you're better off with a terminal block.

[Anonymous No.1243960]

>>1243954
I'm still not sure if they're all being paralleled for current carrying capacity (but I think not), but for low voltage a bus-bar is probably the way to go since it's more rigid than a terminal block and more replaceable/repairable than a mass-solder joint. More expensive than both though.

[Anonymous No.1243965]

>>1243960
>I'm still not sure if they're all being paralleled for current carrying capacity

They are.

And they need to be inserted into the terminal block, that's what the 3D printer has mounted on the control board.

So pretty much carefully twisting them and soldering is about the best I can do?

[Anonymous No.1243967]

>>1243954
>I didn't get to choose their gague
Read the wire, then. See all the repeating printing on the side? It's probably near where it says AWM.
Now, consult this table https://www.powerstream.com/Wire_Size.htm
and find your gauge, particularly the max current for chassis wiring, and the ohms per 1000 ft. Now divide how much your load needs by the current carrying capacity per wire to find out how many wires you need. Also consider the ohms per 1000 ft column and try to avoid significant voltage drop at your required load current by keeping wires short.

>>1243957
So 60 LEDs per meter, thus 20 parallel circuits per meter, so 3.1 * 20 = 62 parallel circuits. Interesting, so if there are 2.33 amps being drawn by 62 parallel circuits, that means they're running at 37mA per circuit, which is indeed within an inch of their life. If instead you have gotten 120 LED/m strips, they would be running at only 18-19mA per circuit, which is slightly more sane.

[Anonymous No.1243968]

>>1243958
My iron is 45W, gonna cross my fingers and give it a try with the iron cranked to 450°C as it's my only option

[Anonymous No.1243973]

>>1243967
I did some measurements with a spare off cut and at 12v there was 2.6v across the on tape 150 ohm resister, so 17.3mA.

These are triple chip LED's btw.

[Anonymous No.1243977]

>>1243973
Oh, okay. Then they're reasonably driven for maximum brightness and within spec.

[Anonymous No.1243978]

>>1243967
>>1243973
I think we are getting confused. 11.5v at LED strip start point, 2.333 amps being drawn.

Now do the numbers add up?

[Anonymous No.1243983]

>>1243978
>>1243977
Ok, 311cm plus or minus a few cm because my room is a shit tip and I can't get to one end.

[Anonymous No.1243986]

>>1243978
Ah, the triple chip part threw me off. So there are three resistors on each section, I presume, and each one is passing ~18mA at 12V for 54mA per section at 12V, so more like 12.5mA each at 11.5V.

[Anonymous No.1243988]

>>1243965
So you have 8 wires from what I can only assume to be multiple power supplies that all need to go into a single one of >>1243954's pic related? Or multiple of the "RAMPS" connectors? Are you planning on combining all those wires and leading a larger gauge wire off them to combine their current capacity, or are you just pulling a same-gauge wire off them instead? Why do you have so many wires in the first place? Can you MS paint something to make it clearer?

[Anonymous No.1243991]

>>1243986
kek, I just came to that conclusion too as the page updated. Also 1 segment is out so its only 2.333 amps/61=38.2mA/3 = 12.7mA per chip string @ 11.5v starting voltage. It would probably be more if there was no tape track resistance.

Lets just assume that the designers are aiming for 20mA per string if 12v was at every section without voltage drop, at 11.65v we are drawing 2.5 amps now. I'm doing this with a 12v SLA so getting 12v exactly at the end of the cable run is impossible.

Also the LED strip is mounted on a metal bed so it technically has a decent heatsink.

[Anonymous No.1243992]

>>1243988
They are all connected to the same power rail in one ATX power supply. Pic related.

They are being twisted and shoved into a terminal like the green one I posted a pic of above (which I mistakenly called a RAMPS connector).

For reference the stock power supply that the printer came with used a single 16AWG wire to carry the entire power load, this is clearly overkill as this PSU is rated for 2x the load of the printer. I was just wondering if there was a better way of connecting the wires than twisting and soldering, but I haven't heard any better ideas so I am just going to do it.

[Anonymous No.1243995]

>>1243992
Don't bother soldering, just fit as many as you can into the screw terminal and tighten it nicely. Add a zip tie around the bundle to keep them from sneaking out and another to the frame for strain relief.

[Anonymous No.1243999]

>>1243992
>Pic related

[Anonymous No.1244000]

>>1243999
Yeah, give it a go in the terminal block since there is nothing to lose. If it gets warm under load then you'll know it needs a better connection.

[Anonymous No.1244003]

>>1243991
Also I guess this shows the disadvantages to extra low voltage lighting systems, a few meters of cable causing half a volt of drop can dictate the lamp current dramatically if a constant power type of driver isn't used.

[Anonymous No.1244004]

>>1243999
Can you crimp a nice ferrule on there?

[Anonymous No.1244006]

>>1243995
Oh cool thanks, I just realized the whole bundle won't fit anyway

[Anonymous No.1244055]

>>1244006
Put them individually to eye connectors, get another thick wire which fits the green terminal and put another eye connector on it, and stack them all together with one bolt. This way you don't have to buy a solid bar of copper.

[Anonymous No.1244056]

Now it's actually Gondola.

[Anonymous No.1244057]

>>1244056
berry bedder :DD

[Anonymous No.1244058]

>ask RS support chat a question
>"its an option for you to decide"
>"its an option..."

Well is sucking my dick also an option?

[Anonymous No.1244060]

>>1244058
pooing in loo is always an option for you to decide

[Anonymous No.1244116]

Hey guys, I am thinking of buying some non functioning honeywell wifi thermostats, but the thing is, doesn't come with the power back
on the left in
>pic related

Does anyone happen to know if I can connect to the pins on on the thermostat itself, and power it that way? I mainly want to observer the circuit board, and possibly part it out.

[Anonymous No.1244118]

>>1244116
I am an idiot, this is the model
>Honeywell TH9320WF5003 Wi-Fi 9000

[Anonymous No.1244120]

>>1244116
Post pics of circuit board, usually it's easy to guess which are power pins

[Anonymous No.1244127]

>>1244120
I haven't bought these yet, I found some on ebay, and figured I would ask first, but regardless, I think I will still buy them, and post pictures when I receive them.

[Anonymous No.1244131]

>>1242947
no

[Anonymous No.1244134]

>>1242969
no
no difference, both are worthless junk

[Anonymous No.1244145]

>>1244116
You can make an educated guess at which pins do what by looking at the circuit itself, clues perhaps on the pcb in the way of silkscreen, big copper pour is likely 0v, on a two side board one side may be a voltage rail. look for power pins on ics, source/drain pins of fets, relays perhaps, check voltage ratings on capacitors can be helpful if you are completely in the dark.
A good current limiting bench supply is very helpful for testing but in the old days an incandescent lamp worked okay.

[Anonymous No.1244150]

can anyone recommend me a cheap-ish bench multimeter?

tired of my shitty handheld and it flopping about everywhere.

[Anonymous No.1244178]

>>1244150

if you cant affix a thing to another thing so it doesnt move around, you dont belong in /diy/

[Anonymous No.1244195]

>>1244178
the whole point of a handheld is being portable, no?
my reasoning was tongue in cheek.

I would have thought it was obvious that i dont mean to replace the handheld, but complement my labs options.
Mainly I want higher accuracy, but also the comfort of no batteries and more features.

[Anonymous No.1244255]

>>1244195
You need at least three, two for normal use and one for reference and system integration. More if OCTED affliction.

[Anonymous No.1244287]

>>1242999
Can you post a link of where you got those?

[Anonymous No.1244292]

>>1243293
>>DS3231 for ultra accurate timekeeping
its the only one that actually keeps time over a longer period, that i have tried.
pic related, keeps time very well.

>>Atmega328 or Attiny2313 based, 20Mhz
doesn't really matter.
any mcu at any clock will be more than fast enough to mux the display and update time from the RTC.

>>555 based boost converter (any better option?)
555 is a bad idea, no regulation.
34063 is a better idea, or buy a cheap module off ebay.

>>Multiplexing with shift registers and transistor arrays
some people claim uln280x wont work, i have tried and it worked fine, but i didn't test for a long period.
reverse breakdown might be an issue.
datasheet states 50v collector emitter, this could be a problem

>>Tilt sensor as a way to set times and alarms by tilting the clock or:
>>VL53L0X range finder for inputting data by waving your hand
oh my god why.

[Anonymous No.1244294]

>>1243377
I'm not shitting on it I'm just saying its the stuff of nightmares. If the rotary has detents it can help but I would prefer a very soft action push button. I'm not saying don't do it, just do it carefully. Or do whatever you want lol none of my fucking business

[Anonymous No.1244301]

>>1244292
forgot pic.

not nixie, filament 7 segment.
had a cpld nixieclock, but the cpld died and i cant be asked to deal with the bloat that is xilinx software again.

[Anonymous No.1244307]

>>1244292
>555 is a bad idea, no regulation.
The memetastic Nixie supply using 555 is regulated. Not very well, but still.

[Anonymous No.1244309]

>>1244307
oh well.
i guess anything is possible.
still 34063 is a better idea.
just as easy and just as cheap as a 555 i'd wager.

[Anonymous No.1244316]

Just started training as an electronics technician. Any tips for me to get ahead? Looking for trade secrets or just general knowledge.

[Anonymous No.1244332]

>>1244316
lurk /ohm/

[Anonymous No.1244336]

>>1244316
depends on what you do.
tell us about your tasks and duties. maybe something will come to mind.

i dont have all my knowlege written down in a list of 'nice to know' sort of way.

[Anonymous No.1244343]

>>1243531
Sorry for not answering I did not forget about you, I've been doing some testing. It might take a while.
>>1244301
10/10 clock, would aloha a snack bar
>34063 is a better idea
Neat, I didn't know those existed, I have to try them out, I just realized there are hundreds of nixie PSU schematics using that IC, it's also cheap, thanks!
The 555's are regulated, but they're finicky since they are not designed for that purpose and require many components to work properly. You just saved me a headache.
>but the cpld died and i cant be asked to deal with the bloat that is xilinx software again.
Sorry man, that sucks.

>>1243377
>>1243334
I don't want anything protruding out of the case and altering the minimal and simple aesthetic. Everything will be flush. Also >>1243347, rotary encoders give me hell, why can't they output HIGH on one side when turning right and HIGH on the other side when turning left? Why do they force me to use interrupts and shit to be able to tell which side has turned HIGH a few milliseconds faster than the other, fuuuu?
>Radio time receiver is also a good idea.
With the DS3231 RTC I can forget about setting the time for a long time, it is accurate to +-2 minutes per year.

>>1244316
https://imgur.com/a/21Jz9
https://imgur.com/a/YC0Du
Here are some pictures a seasoned technician posted on leddit some time ago:
https://www.reddit.com/r/electronics/comments/6wqamc/im_an_appliance_repair_tech_here_is_an_annotated/

[Anonymous No.1244345]

>>1244336
Just started 2 weeks ago at post secondary and will most likely be going into telecommunications

[Anonymous No.1244349]

>>1244343
>10/10 clock, would aloha a snack bar
thanks, my dude.
>You just saved me a headache.
my pleasure, there are a mountain of cheap useful chips out there and I probably only know a few of them.

>Sorry man, that sucks.
its ok, it kept terrible time anyway.

what is worse is that i have a big pile of nixies like yours.
two of them are different in particular, one shows prefixes(µ,m,k,M ect) and the other shows units (A,Ω,V,Hz,W,F etc).

I wanted to make a bench multimeter using this display, but i realized fast that it would never be as accurate as a real multimeter.
I guess i could hope to retrofit the display onto an existing unit, but that sounds like a headache for no real gain other than aesthetics.

like pic related, but vertical and oval like yours.

[Anonymous No.1244350]

>>1244345
>telecommunications
i cannot help you.

[Anonymous No.1244351]

>>1244349
my most recent use of a nixie.
its just slaved to the relay drivers for channel indication.

[Anonymous No.1244352]

>>1243540
>10 per tube
Hope you are using stencil and reflow oven, with base resistors that is 300 pads to paste, solder and check.

[Anonymous No.1244356]

>>1244345
hmmm maybe look into ham radio? idk.

[Anonymous No.1244358]

>>1244134
You'd probably say the same thing about my $5 multimeter, but it's just about good enough for me. Those things have more features than a low frequency analogue scope, so I'll probably go for one since they're cheaper by a factor of 5 at least.

>>1244349
>IN-19B
Very nice, though I can't imagine you'll be using the Tesla unit on them very often. Got any IN-18s to go with it?

[Anonymous No.1244359]

>>1244358

nah dude.
the unit/prefix tubes are like this guys tubes>>1244343
just didn't have a pic, so i took one off google.
i have no idea what those tubes are called.

i got plenty of regular ones in that format as well, and sockets on an old board.
just never got around to doing anything with them.

distracted by other projects for years now.

[Anonymous No.1244365]

>>1244352
I'm driving the transistors with cd4017s so I only need one resistor per tube. It's not that bad desu.

[Anonymous No.1244374]

>>1244294
>rotary has detents
Most of the cheap ones do.
>I would prefer a very soft action push button
Most of the cheap ones with built-in push switches have semi-soft tactile action, just not very soft. Maybe the feel is different with a knob mounted on them.

>>1244301
>cpld died
Quartus is surprisingly slightly less terrible and Altera chips are easy to find.

>>1244343
>I don't want anything protruding out of the case and altering the minimal and simple aesthetic.
Then you probably want touch sensors, for a few burgers a dozen from your favorite aliexpress seller.
>rotary encoders give me hell
Obviously you use just one interrupt on one line which then immediately samples the other line just once to determine the direction of rotation.
>interrupts and shit
If you want to make anything that responds to the real world, you better get used to it, kid.
>technician = repairman
Lurk more.

[Anonymous No.1244427]

How come Rigol isn't trash, despite being Chinese?

[Anonymous No.1244457]

>>1244427
Maybe because they use it themselves. Working technicians can't afford to be stalled or misled by shit. Also, it's a higher-tickee product than random OTT TV box of the month.

[Anonymous No.1244460]

anyone built a quadcopter? looking for part recommendations for a small-medium sized one

[Anonymous No.1244461]

These breadboards are top-tier when you need to make a little discrete building block like a bidirectional level shifter or something.

[Anonymous No.1244462]

>>1244460
That's more mechanical than electronic. You might be better off starting a thread on the topic.

[Anonymous No.1244532]

>>1244460
>>1244462
We have a general for that shit.
>>1241106

[Anonymous No.1244551]

>>1244461
>These breadboards are top-tier

holy fuck. people actually use suface mount parts by choice? (meaning handmade circuits).

[Anonymous No.1244552]

>>1244551
There are many components that only come in SOIC or SOT footprints, and they're often cheaper anyways. If you're making a small PCB they can't be beaten.

[Anonymous No.1244554]

Hey guys simple question,

I need to buy a wire for 36 VAC 8.333A for 20 meter in a wet room that is about 15 degrees Celsius. The wire is going to be moved regularly, so I copper wiring.

Is 2.5mm copper wire good enough? I also have some 1.5mm btw.

[Anonymous No.1244564]

>>1244551
You should try solder paste and hot air reflow some time. Not necessarily for a little circuit like this, but if you have a lot of connections to solder, the experience can be downright magical.

[Anonymous No.1244565]

>>1244554
hmm I've used a calculator and it says I need to use 21 mm^2 or 5.2 mm, is this right? this sounds like a lot.

oh
>2.5mm
2.5mm^2
>1.5mm
1.5mm^2

my bad

[Anonymous No.1244589]

>>1244565
>>1244554
Dude, what?

[Anonymous No.1244644]

>>1244589
I want to know the area of the cable that I need to buy.
I think I'm going to use a copper cable with a 2.5mm^2 (mm squared) area.
Some online calculators say 2.5 is fine, others say it's way too small.

I don't have any specific numbers, but I want to minimise power loss, so if I would gain a few percentages in power by using a thicker wire I'll just buy a better cable.

[Anonymous No.1244645]

>>1244554
Where about are you?
Mains voltage 1.5mm2 is max you can put in a 13A plug should give you an idea, 2.5mm2 radial is something like 20A breaker if my memory serves
Of course voltage drop will become important at extra low voltage but I think hey 20m isn't crazy long, 36v isn't nothing to be sniffed at either.
Give it a try if I were you, if you have the cable I would try 1.5mm2 it's not going to catch fire or anything, measure voltage under load see what happens.

Very concerned that calculator said no way, problem somewhere either with the calculator or your numbers which is a concern...

[Anonymous No.1244647]

>>1244644
>>1244645
Ok volt drop might be a concern, 3 or 4 volts at 1.5, 2ish volts at 2.5. 4mm wet old be fine

[Anonymous No.1244650]

>>1242891

Can I use a chinkshit high current SCR dimmer to regulate inductive load, i.e. the transformers in my nigger rigged MOT welder, for when I want to weld thin metal? Variacs are expensive as fuck.
Interference is probably not an issue here since it's completely encased in aluminum, what other possible issues are there?

[Anonymous No.1244653]

>>1244644
it all depends on what powerloss is acceptable. for -3V you're good with 1.5mm2 (with a loss of 30 watts on the line) for -1v you need almost 4mm2
the low voltage is fucking you, 3 volts on the line are almost 10 percent loss
But get a propper isulation, as that matters more. You want high flexible leads and something like rubber or PUR for the isolation

[Anonymous No.1244715]

TL:DR cool, complex project idea? if you could build whatever you wanted what would you build?

Hey all, am studying engineering at uni and for my second year we get to build a project, pretty much anything we want, so I want to build something cool since we got a year to do it. Its general engineering this year but next year I get to do electronic hence I want to do an electronic project.

I've googled around but most projects seem to be really simplistic even apparently ones for final year students... probably because most fuckers will throw them together last weekend before due date to get a measly pass and be content in the mediocrity. That probably sounded pretty fedory but I just really want to be good at what I do. I can build the actual project, a model of it or even just a research paper on a subject. I want to sink a lot of hours into it. The more obscure the better since i usually find things like that more interesting.

Am not asking for any guidance or anything btw, just throw the name of something at me. Thanks.

[Anonymous No.1244725]

>>1244715
Build a small scale SkyTrain using LIM motors

[Anonymous No.1244731]

>>1244715
an autonomous rover that can map the interior of a house. Look into project Tango or other SLAM technologies.

A device that can capture your brain activity and convert it into useful output, e.g. figure out when you're thinking "move THAT object", or "jump!" and make a robot or a virtual agent follow those commands.

Build a high power variable frequency drive with circuit protection and all the trimmings.

Build the highest quality audio amplifier with stupid gimmicks like moving vacuum tubes and sell it to audiophile hipsters for $50,000.

[Anonymous No.1244764]

>>1244731
And have the actual signal path be high-quality solid state parts.

[Anonymous No.1244819]

>>1244731
>A device that can capture your brain activity

[Anonymous No.1244830]

>>1244819
>what has already been invented can never be recreated
I wonder what we're all doing on this board

[Anonymous No.1244840]

I'm interested in building this kind of bus-controlled lighting control system (pic related). I wrote a long post but unfortunately 4chan thinks it's spam.

Therefore I had to copy it to pastebin: https://pastebin.com/zT3Tin5J

Any comments about it are welcome.

[Anonymous No.1244864]

So after a lot of reading and some testing of my own, boost converters in parallel don't play so nicely with each other, but what about boost converters in series? What matters more for efficiency, is it the output voltage or the current?
I'm trying to step up 3.7v to 12v with 800mAh current and have a lot of 10 mt3608 boards to fiddle around with. They were cheap and seemed fine for my project.

[Anonymous No.1244866]

>>1244840
If you want a bus to go all round your house you want to consider using current based signaling rather than voltage based. Limitation is that you can't split without repeaters because Kirchhoff you know?
Check out rs485, pretty sure that's current based and its differentially signaled to remove common mode noise.

Pulling down a tied bus like you say is pretty common especially where you have multiple transmitters contending for space on the bus, can't have it driven both ways, not good at all. Also good for detecting collisions, if you aren't driving the bus but you see it is driven then you have collided, makes priority messaging a cake walk too this way.
It's a good technique but might not be up to the physically

[Anonymous No.1244869]

>>1244554
You didn't mention how much voltage drop you can tolerate on the far end. Anyway, your 2.5mm wire will drop roughly 2/3 V each way.

>>1244864
>in series
You might want to think that through a little more. Like, where's your ground for the other converters?

>>1244650
I doubt it.

[Anonymous No.1244874]

When dealing with transformers, does anything really matter except the ratio of turns between primary and secondary?

Like say I have a 240v to 24v transformer, if I feed it 80v, will I get 8 out or will it even work at all?

[Anonymous No.1244888]

>>1244874
The winding resistances are chosen so that the voltage drop is acceptable at 24V output. The voltage drop at the nominal current might not be acceptable at 8V output.
Also,mains transformers saturate quickly if you try to run them at higher than rated voltage.

[Anonymous No.1244890]

>>1244874

ignore the windbag, the answer is yes: the turns ratio works at any voltage. and also in either direction: if you put in 8V on the secondary, you'll get 10x that on the primary.

[Anonymous No.1244893]

>>1244890
>there is no such thing as a non-ideal transformer

[Anonymous No.1244903]

>>1244650
perhaps you could compensate for the phase shift with a capacitor?

[Anonymous No.1244907]

>>1244650
if it's a leading edge dimmer. if it's a falling edge dimmer you'll be very sad. i'm not a welder but it'll also impact your weld properties.

[Anonymous No.1244937]

>>1244907

Is there a way to tell, if there's no diagram and I don't have an oscilloscope?
I only tried it with a scavenged 220V fan so far to see if it works at all. Seemed to work fine, except obviously the fan didn't start to spin until the voltage hit some ~180V.
Well, guess I'll put it together out of the box and try welding for a while, if it works as intended and doesn't burn out, then I'll build it into the box and call it a day.

[Anonymous No.1244941]

>>1244937
you'll know right away what the difference is, and most dimmers are leading edge. SCR dimmers are by definition but i sure as shit wouldn't trust an ebay description. you're probably fine.

[Anonymous No.1244952]

>>1244830
>we're all
Dead giveaway. You're talking about your own problems, Mr. Straw. Do you have to invent the wheel to use it as a flywheel? Do you have to invent that EEG-type headset to use it for a different purpose?

[Anonymous No.1245057]

>>1244551
0805 passives and SOIC are ezpz

[Anonymous No.1245102]

>>1245057
Even fine-pitch QFPs are easy: use a hoof tip, or solder paste and a wedge tip, and use tacky flux to hold it in place.
https://www.youtube.com/watch?v=nyele3CIs-U

[Anonymous No.1245110]

>>1244874
Generally you can feed the primary with less voltage and be fine.
Using the secondary as the primary can be trickier, you just need to make sure that neither winding has it's current limit exceeded.

[Anonymous No.1245111]

>>1244840
Have you thought about using arduino ?

[Anonymous No.1245112]

>>1244864
>boost converters in series
You don't need two stages to go from lithium to 12V

[Anonymous No.1245495]

So im replacing some gamesave batteries in my old Pokemon gameboy cartridges.
I get to this one, open it up and the battery has leaked or something. I snipped the battery out.

Whats the best way to clean it?
Scrub it with Isopropyl alcohol, install the new battery and hope it turns on?

[Anonymous No.1245500]

>>1245495
Yeah, that is a good solution to use. Either it works or it doesn't.

[Anonymous No.1245547]

I need a barometer sensor (preferably digital) that can go down to 10Pa with a decent accuracy of ±2Pa or so, do these exist? What about 100Pa ±20?

[Anonymous No.1245552]

>>1245547
>barometer
>can go down to 10Pa
As in, 10Pa from full vacuum? Or differential pressure in that range?

[Anonymous No.1245571]

>>1245552
Never mind, I didn't expect part sites to have an individual category for pressure sensors with filterable specs.

[Anonymous No.1245580]

>>1245495
>I get to this one, open it up and the battery has leaked or something

that looks more like 'left in a roadside ditch'

[Anonymous No.1245627]

>>1242891
High pin count schematics are a pain. Is there a quick way to create a component in kicad with 200 pins?

[Anonymous No.1245633]

>>1245495
before soldering a new battery you should test if it actually works by powering externally

[Anonymous No.1245749]

>>1245495
it should turn on either way, it just wont retain saves if there is no battery.

clean first either way.

[Anonymous No.1245843]

>it should turn on either way

batteries tend to have acidic (or basic) liquids inside which can eat up copper traces, or the pins on lil chips.

[Anonymous No.1245848]

>>1245843
if it doesnt turn on, throw it away

[Anonymous No.1245880]

>>1245627
Components: Not really. I ended up writing a throwaway script to create a .lib fragment. If you have a csv file to drive the process, it may be easier.
Footprint: Much easier. Create one pad, then right-click and create pad array.

[Anonymous No.1245907]

Does anyone know what that thing at I3 is?

[Anonymous No.1245910]

>>1245907
current mirror?

[Anonymous No.1245914]

>>1245907
>>1245910
yep, current mirror

[Anonymous No.1245923]

>>1245910
>>1245914
Thanks

[Anonymous No.1245928]

>>1245907
outdated symbol for current source, I3=1mA
a current mirror cannot be a two-pole

[Anonymous No.1245946]

>>1245928
how did you get 1mA?

[Anonymous No.1245951]

>>1245946
Assuming the -8V is a measuring point, the current through the collector current is 2V/2Kohm = 1 mA
the base current is less than 0.1 mA
so the emmiter current woud have to be ~0.9 mA?

[Anonymous No.1245955]

>>1245951
base current is 1mA/B and B is 100..600 for small signal transistors

[Anonymous No.1245961]

so today I learned that sometimes chinks will just say fuck it to putting a ballast in their flashlights and if you use rechargeable or high current batteries instead of alkalines your flashlight will burn up

I scraped off the dead diodes and soldered on some new ones

it works but it's not as bright as it was before so I'm still mad

[Anonymous No.1245966]

>>1245951
B is defined as Ic/Ib and Ie=Ic+Ib which means I3=Ie=1mA+1mA/B and V4=1V-10kΩ*1mA/B
>>1245961
Did you add a small series resistor to prevent this malheur from happening again? Almost all cheap 3xAAA flashlights rely on the internal resistance of the batteries.

[Anonymous No.1245982]

>>1245848
It could be repaired, those traces look fairly generous.

[Anonymous No.1245983]

>>1245928
Outdated? That's what we were taught, though with the arrow on the side.

[Anonymous No.1245999]

>>1245982
sure, it also seems to be missing some parts, but also it looks like it never had them in the first place.

U4 could be fucked, I dont know if replacements are any easy to get, it seems to be some battery related device

[Anonymous No.1246000]

>>1245999
crack open another cartridge you don't care about?

[Anonymous No.1246009]

>>1245999
Im the guy with the cart, I bought it new and it worked like 5 years ago so no parts are missing.

I have a handful of others to finish replacing the batteries first before I try to really clean this up. The picture doesnt do it justice, its nasty

[Anonymous No.1246015]

>>1245983
different conventions I guess

[Anonymous No.1246021]

how do I work out how long it'll take my device to charge? cable is 0.5a battery is 3,300mAh want to set a timer on the switch so my battery only charges to 100%

[Anonymous No.1246023]

>>1245999
MM1134s appear to be discontinued.
MAX818 appears to be a functional replacement, though not pin compatible.

>>1246021
Ordinarily you divide amp-hours by amps to get estimated hours, but in practice this will only get you to 80-90%. Only experimentation will tell you how long the constant-voltage phase of the charge will last, as it varies with aging and between units.

[Anonymous No.1246044]

>>1246021
>set a timer on the switch so my battery only charges to 100%

there's no answer to this question unless you know how low the battery is at the start of the charge cycle. going from 90%-100% is gonna be a fast, whereas 10-100 is gonna be slow.

that's why you need a proper monitoring circuit that will stop charge when full. LiPos can explode otherwise, whereas nicads and nimhs will survive but will lose longevity.

[Anonymous No.1246049]

>>1246044
>proper monitoring circuit
is this something I can just buy for my phone?

[Anonymous No.1246053]

>>1246049
>for my phone?
If it's a cell phone it has monitoring of the battery built-in.
It doesn't turn the charger off but it internally sets the charge rate to a maintenance level.

[Anonymous No.1246054]

>>1246053
so charging over night isn't something I should worry about?

[Anonymous No.1246057]

>>1246054
If you are using a 'standard' USB style charger that outputs 5vdc you can leave the phone connected indefinitely.

Is this your first phone?

[Anonymous No.1246059]

>>1246057
yeah just a standard usb type c plugged into a PS4, my first smartphone yes the harder to replace battery is worrying me so I want to prolong it

[Anonymous No.1246094]

>>1245112
>You don't need two stages
True, but you probably don't get 800mA from a single MT3608 at 3V input.
>>1244864
Connect a test load resistor (e.g. 33Ω 5W, will get hot) and carefully adjust the modules to the same output voltage (zero difference). Connect another 33Ω 5W in parallel and see if you still have about 12V. Let it run for a while and check the zero adjustment.

[Anonymous No.1246139]

>>1246094
What if you desolder a potentiometer from one of the modules and then connect one potentiometer to both modules? Will it work?

[Anonymous No.1246149]

Hey /diy/
Im troubleshooting a SF466A-24 SMPS.
It does have multiple DC outputs, however the 3.3V one is not working properly.
Problem is, at Startup the 3.3V is arround 3V but noisy as fuck, as the Logic level needs the 3.3V it doesnt start the device. When i heat it up with an hair dryer, the Signal is getting less noisy untill it is clear at 3.6V and the device starts.
The board has no signs of visual damage

Can this issue be related to bad caps? this power supply is famous for having shit caps, but i already replaced them 3 years ago

[Anonymous No.1246161]

>>1246149
Yes, if your replacement caps were bad or not low-ESR rated then it can be bad caps. Pic related is great for testing caps, you can find it on aliexpress by searching for "esr meter".

[Anonymous No.1246235]

>>1246139
Would be nice, wouldn't it. Sadly, tolerances prevent that.

[Anonymous No.1246257]

>>1246149
>3.3V one is not working properly
both 3.3V outputs or only one? Thermal effect can mean bad joint. I would resolder anything in this area.

[Anonymous No.1246262]

There's an attic antenna that has a cable like pic related coming down into my room. What'd be the easiest way to extend the cable and connect it to my television?

pardon if this isn't the right thread.

[Anonymous No.1246272]

>>1246262
Nevermind, it was actually as fucking easy as I thought it would.

I had one of these laying around, so I connected the twin leads to it, shoved a threaded in-line connector in there, and ran a standard coax cable between the in-line and my TV and it worked.

Huzzah free TV.

[Anonymous No.1246276]

I have a mutlimeter, how do I find out the ampage of a cable?

[Anonymous No.1246279]

>>1246257
yeah, will do that. can you tell me where you got this pic from? ive been searching all day for documentation for the power supply
the 3.3L is somehow a little bitch, as its a standby voltage. If i recall it was 0 but i dont know if this might be intentionall while its already trying to boot

[Anonymous No.1246284]

>>1246276
you have to select the correct ohmage setting on the multagemeterage, and then justage place theage leadsage atage theage extremitiesage ofage theage cableage

[Anonymous No.1246285]

>>1246279
sorry

[Anonymous No.1246288]

>>1246285
try adding spaces around the symbols

[Anonymous No.1246291]

>>1246285
i just tiped in the url and did work, ty!

[Anonymous No.1246295]

>>1246276
>ampage
Do you mean ampacity?

[Anonymous No.1246312]

>>1246295
If I didn't mean amperage I wouldn't have said amperage.

[Anonymous No.1246347]

what sounds like a typical range of resistance for the mains-end winding of a transformer in an old, really simple 13.8V 50W power supply? I'm trying to fix my broken one that's blowing fuses. The only transistor still has its voltage drops in the appropriate range but the primary winding is reading like 4 ohms and that sounds really really low to me

[Anonymous No.1246354]

>>1246312
>>1246295
why can english separate the unit from the phenomenon anymore?

used to be tension/excitement measured in volts, not voltage.

its current, measured in ampere.
not amperage and ampacity is retarded.

its as stupid as saying OHMage instead of resistance
wattage is also stupid.

[Anonymous No.1246355]

>>1246347
i've got a 240W transformer with a 5.6 ohmage primary and a 48W transformer with a 29 ohmage primary. the 240W might be 2.8 ohmages because there's 2 primaries in parallel and i can't remember whether i measured one or both. sounds like your transformer has had some winding enamel wear or melt off and cause a partial short. if it's not soldered into the supply you should take it out and test the output voltage with a small signal on the primary. if it's lower than expected that'll line up with a short.

[Anonymous No.1246359]

>>1246355
upon closer inspection it's reading 1.8 when I dig past the oxidation on the terminals, and there are pretty obvious signs of overheating/burning, which I just mistook for age. Everything south of the transformer is relatively clean and bright.

Is there an easy way to determine what the output voltage of the trans is SUPPOSED to be? I might just try re-winding it since it's old enough to just have some lacquer/wax holding it together instead of being covered in robot jizz.

[Anonymous No.1246361]

>>1246355
>48W transformer with a 29 ohmage primary
I have a 40w transformer with a 13ohm primary.

[Anonymous No.1246370]

>>1246361
add a 16 ohm resistor in series to correct your underohmage. i wouldn't be surprised if my shitbrick is wrong though.

>>1246359
anything under 12-16vac (or equivalent ratio to input) would be obviously wrong.

[Anonymous No.1246374]

Good night /ohm/. Remember to always store your webers in a dry and well-ventilated place.

[Anonymous No.1246376]

>>1246347
4 ohms is reasonable. An ideal transformer has zero ohmage.

[Anonymous No.1246378]

>>1246361
>>1246376
>ohmage.

am i being trolled?

[Anonymous No.1246386]

>>1246376
i got interested and did some napkin math on my transformer.

core area: 497mm^2
winding length: 127mm
winding area: 433mm^2

120V/(4.4*50Hz*(497mm^2)*1T) = 1100 turns
433mm^2/1100 = 0.4mm^2 -> 28awg (0.2ohms/m)
127mm*1100*0.2ohms/m = 28 ohms

if you run the numbers for 2 tesla you get 6 ohms. so i guess the full range is reasonable depending on the steel. anon's transformer will probably be rated for 75W+ anyway with a 50W output.

[Anonymous No.1246412]

>>1246378
yes

[Anonymous No.1246560]

>>1246354
>ampacity
That's something else entirely. Ampacity is defined as the maximum amount of electric current a conductor or device can carry before sustaining immediate or progressive deterioration.

[Anonymous No.1246597]

>>1246560
also known as current carrying capacity in first world countries

[Anonyrnous No.1246616]

So newb here. What circuit should I use to accept any voltage between 1V and 5V but drop anything between below 1V? I want to work with the 2-5V range but reject anything below 2V.
Tried using a Schmitt-Trigger but voltage was either on or off.

[Anonymous No.1246620]

>>1246616
Define "drop".

[Anonyrnous No.1246622]

>>1246620
Drop as in not send it into the circuit. Basically I want to accept any voltage 1V-5V but when the threshold is below 1V, I want it to read 0V.

[Anonymous No.1246625]

>>1246597
It's informal, but still a more widely used and is officially defined by National Electrical Safety Codes, in some North American countries.

[Anonymous No.1246628]

>>1246622
Okay, I asked because you have to send something into the circuit. 0V is a valid answer.
Quick and dirty: An open-collector comparator like the LM393 could be used to test for the cutoff point. Also send the input voltage through a high value resistor to a voltage follower, and connect the comparator output to the voltage follower input on the non-input side of the resistor.

[Anonyrnous No.1246640]

>>1246628
>LM393
Interesting. Thanks.
I'll do some research on this and try to design a schematic.

[Anonymous No.1246646]

>>1246616
>What circuit should I use to accept any voltage between 1V and 5V but drop anything between below 1V? I want to work with the 2-5V range but reject anything below 2V.

these are contradictory requirements. the first requirement can be accomplished with a circuit like this. when input voltage (called V- on schematic for some dumb reason) goes above 1V (or whatever voltage you put at pin 5 by choosing a value for R1) then the comparator flips low, which is then inverted high, which then turns on a 4016 analog switch which lets the input voltage flow out to whatever is waiting.

R3 provides hysteresis, and can be adjusted to taste.

if the 4016 somehow interferes with the signal, then it can be replaced by a tiny reed relay.

[Anonymous No.1246647]

>>1246616
"accept any voltage between 1V and 5V.."
"..reject anything below 2V."
Requires advanced magictronics.
Tell your machine to if..then..else.

[Anonymous No.1246652]

>>1246646
not him but what is the use of hysteresis in this case?

[Anonymous No.1246657]

>>1246652
The usual: to prevent obnoxious chattering when the input voltage is near to the limit voltage.

[Anonymous No.1246719]

Any tips on reducing acoustic noise of inductors? In SMPS I used smd inductor with very little noise, but same design with through hole inductor and shit can be heard across the room.

[Anonymous No.1246732]

>>1246719
Power save mode? Add load to check. Epoxy the inductor. Replace leaded inductor with SMD inductor and hear what happens.

[Anonyrnous No.1246738]

>>1246647
The input is 0V-5V, the load should only accept an input voltage above 1V. (I said initially 2V but disregard that).

[Anonyrnous No.1246743]

>>1246646
Got it. So nothing happens when voltage is over reference. When it's below, the comparator shuts off the relay/switch.

[Anonymous No.1246765]

Can I replace the one on the left with the one on the right?

It's going in a ceiling fan.

>Get the right one

There are none for sale in my country and you are lucky to scavenge a capacitor.

[Anonymous No.1246771]

>>1246719
- some inductors are noisy due to loose windings -> add glue
- some inductors are noisy due to the core material -> use inductor with different core material
- inductors get noisier when you drive them closer to saturation -> use inductor with higher saturation current

Like the other anon said, low power load can make the switcher to skip pulses, resulting in audible noise. Current mode switchers are also prone to subharmonic oscillations.

[Anonymous No.1246789]

>>1246765

Image would help

[Anonymous No.1246810]

>>1246789
Pretty new to electronics and all, but it seems it would depend on what the purpose of the capacitor is.

[Anonymous No.1246814]

>>1246810

I am not sure either. But fans with bad capacitors seem to start to slow down or refuse to start without a helping hand. Mine are slowing down

[Anonymous No.1246822]

>>1246814
After a bit of digging, the answer is apparently no. You have to replace it with a cap of an equal capacitance.

[Anonymous No.1246824]

>a few years back
>be edgy gaymen kid
>wow those new mechanical keyboards are nice
>buy steelseries 6gv2
>it is nice
>now a few years later
>it is still nice
>those gaymen nigger cunts designed it that there is no windows key on the left side
>annoys the fuck outta me
>too greedy to buy a new one
>we gotta fix this
>open
>look on the back
>wow this looks easy
Just gotta put some wires across and hope that it will just map the windows key to the left side. Will see if this works.
The only problems is that there are 4 contacts on the right windows key, but I think I know which ones I need to pick.

Progress report in a few minutes. Gotta make myself a meal first

[Anonymous No.1246827]

>>1246824
FUCK. forgot picture

[Anonymous No.1246837]

>>1246827
>silkscreen says win-l
>clearly has pads and connections
I dont think you should do this mang.
the button probably does something, just not registered to win.
get swiftkeys or something and remap it to win-L

[Anonymous No.1246841]

>>1246837
No it does nothing. Already looked at it. It is hard mapped to the multimedia keys on the top row, which can only be pressed if this is pressed.
So I will try if I can then still use the multimedia keys afterwards

[Anonymous No.1246842]

Got it done.
Windows key on the left is working now.
Multimedia keys aint workin though. Next pic with some more thoughts comin in

[Anonymous No.1246843]

>>1246842
Now gotta cut those two traces and add another key to get those multimedia keys working again. Will think if I can get this workin

[Anonymous No.1246845]

>>1246822

Sigh. Annoying.

Thank you.

[Anonymous No.1246850]

What is the proper way to connect a DUT to a vector network analyzer?

My Agilent E5100A has two RF out ports with an internal power splitter (option 002) and two receiver ports R and A (I don't have the option that adds ports B and C)

I assume at least one RF out and one receive port needs to be connected to the DUT but some reference material online shows the other RF out and receive port connected as well. From what I gather if you only connect one TX and RX port then you can only get the S11 and the S12 parameters and to get the full complement of S-parameters, S11, S12, S21, S22 you need both. Some diagrams show RF port 1 connected to the DUT and the DUT connected to receiver port A while RF port 2 and receiver port R are bridged. Other diagrams show both receiver ports connected to the DUT with one RF port connected.

I don't really know.

[Anonymous No.1246851]

>>1246765
>>1246789
I'm gonna say yes it should be alright. Same voltage so that's no problem and the new one has a 25% higher capacitance, which seems like a bit much but all it's doing is increasing the phase angle between the startup and main coil.

Try it and if you die let us know.

[Anonymous No.1246853]

>>1246765
>replace the one on the left with the one on the right?
>>1246789
1.75mf and 2.2mf
You can but the speed will be wrong

[Anonymous No.1246855]

>>1246814
>Mine are slowing down
does it turn freely by hand?
bad bearings will make it slow down

[Anonymous No.1246860]

Can someone tell me if I did this right? Or should I connect both antenna connectors to the Yagi element? It doesn't seem to want to stay connected to any routers I point it at.

[Anonymous No.1246867]

>>1246850
ok anon as a university phd in microwaves dropout who broke a 95000£ VNA i feel absolutely equipped to fully give you a half assed answer.

the R input is reference, this has to connect to one of the RF OUT ports, the other rf port goes into your DUT to stimulate.
i guess they are connected, this will give you S11, the reflected power from that one port of your DUT for S12, the transmitted power through a 2 port device like a filter you connect the output of your dut to port A and measure there.
for S22, S21 i guess you have to strip it down and flip it around or something?

like i said i dropped out...

so for you saying rf1->dut/dut->a, rf2->R i would say would give you S11, S12
in your "other diagrams" i assume your rf port is tapped into the link between R and DUT which case its the same really except the connection is outside instead of inside the machine.

[Anonymous No.1246871]

So I was looking into some projects to use a few arduinos I have sitting around for and I have a few ideas just need to narrow down how to use them.
The first and main idea is there a way to create a spherical screen like for a monitor? I want to take a spherical screen, hide all the arduino parts inside with just minimal parts on the bottom exposed for charging, but I want to project a map of the world on it with the wind currents from https://earth.nullschool.net/
Another idea for me is using the arduino to act as a server, which I know how to do this one, I just can't think of what to use the server for.
So does anyone have any idea if the first idea is possible and also if you have a server what do you use it for?

[Anonymous No.1246872]

>>1246851

>Try it and if you die let us know

Sure thing

>>1246853

>You can but the speed will be wrong

I will probably keep an eye out at the landfills and hope to come across something or the next time I go abroad I visit and electronics shop and bring them in. I haven't seen a capacitor on the shelves in years.

>>1246855

Spin real free.

[Anonymous No.1246881]

>>1246872
amerifat here
I do maintenance work and have several dead fans just sitting her.
I'd love to send you a cap but since we use 120vac, all the caps I have are 250v or 270v.
Good luck in finding one. :thumbsup:

[Anonymous No.1246900]

/g/ wasn't helpful so I guess I'll ask here

I just picked up an STM32F429 discovery board and am looking to program it. There are like 15 different tutorials all using different software and telling me to do different things, and the official documentation suggests using 5000 dollar software tools.

Is there a good site to git gud at more advanced ARM-cortex embedded stuff, or should I just pirate one of the professional tools and learn there?

[Anonymous No.1246909]

>>1246867
I'm itching to know how you broke it. Just too much input power? Static discharge?

[Anonymous No.1246911]

>>1246909
nothing so interesting, just unplugged it while it was on. it ran windows embedded and threw its toys out of the pram so to speak. had to be sent back for repairs.

[Anonymous No.1246916]

>>1246860
critical part not visible

[Anonymous No.1246941]

>>1246871
If YOU can't break the problem into its constituent parts and know whether or not you can fit them into your design constraints, how do YOU expect to complete the entire project?
"If you have to ask, no, YOU can't do it."

>>1246900
Just pick the tutorial that uses the tools and loaders that are most conveniently handy to you, and favor those that use gcc instead of one of the proprietary compilers. If you don't get anywhere, pick another one. Eventually you'll get some code built and running on the device, which probably won't change much between workflows. Now add weight and do your reps.
fwiw, STM32Cube seems to be the officially supported toolchain.

[Anonymous No.1246968]

>>1246916
I don't have a pic of it before I glued it up but it's hooked up like this, EXCEPT as you can see, the coax feeding it is to the side of the balun loop instead of right in the middle... but it still contacts the dipole element right at the end . It was kind of tricky to solder so I HOPE I did it right. Do both antennas on the router need to be connected to it?

[Anonymous No.1247011]

>>1246968
>Do both antennas on the router need to be connected to it?
Just put the wire whip on the other side.

[Anonymous No.1247055]

>>1247011
Other side of the router? I've tried hooking it up to each side separately, same result.

[Anonymous No.1247097]

>>1247055
check your router info for how they are using the antennas

[Anonymous No.1247109]

How can I find out the capacity of an SMD ceramic capacitor? It has no markings.

[Anonymous No.1247114]

>>1247109
Capacitor meter. Failing that, reverse engineering.

[Anonymous No.1247115]

>>1247109
i assume you don't have a scope or a multimeter with a capacitance range. if you have an arduino (or anything with an adc) and a high value resistor you can measure the resistor and then set up a circuit like this with a few lines of code.

[Anonymous No.1247191]

>>1243137 here

I've got a small circuit that somehow does what I wanted, but I'm having trouble with the positive voltage, here's my problem:
Vin is between -1V and +1V, Vout should be -3V when Vin is -1, then grow steadily to +3V until Vin is 0V and then stay at 3V for anything above 0V (until it drops again).
Everything is fine for -1V to 0V, but for a positive voltage it continues to grow to 9V. How can I "make it stop" at 3V? I've tried it with a transmission gate, but I've got problems with that as well.

[Anonymous No.1247193]

>>1247191
Simple dumb solution: use a rail-to-rail opamp and give it +3V supply voltage.
Or use an (ideal) diode clipper at output.

[Anonymous No.1247196]

>>1247193
Yeah, I just figured the solution with the op amp.
But wouldn't that be a problem, if the +-3V op amp is half of the time at its maximum? And feeding it 9V?
Gonna try out the diode clipper next!
Thanks a bunch!

[Anonymous No.1247233]

>>1247191
>grow steadily
not possible with positive feedback

[Anonymous No.1247243]

>>1247097
Oh shit, thanks!

[Anonymous No.1247290]

>>1247196
>the diode clipper

[Anonymous No.1247295]

Hope I ill get an answer before this thread is on auto sage.
I have a project in which I'm using a raspberry pi zero and an arduino uno.
I want to power them from the same 5v rail. This 5v line is coming from a step down converter.
My question: what is the most stable way to connect them? If I connect them parallel wouldn't the 5v divide between both components? I know the raspberry pi can power the arduino from its 5v pin, but its the same line as the 5v main rail, so that's essentially parallel as well.

[Anonymous No.1247296]

>>1247295
What instability do you expect if you supply them both from the same source?

[Anonymous No.1247307]

>>1247295
connect them in parallel they both get 5v
you divide voltage among them when you connect in series

[Anonymous No.1247314]

Also remember RPi GPIO pins are 3.3v

[Anonymous No.1247320]

>>1246900
I was in the same position a while back. I got an stm32f103c8xx to get familiar with some arm. After going through the internet I settled on the chink ST-link2 which is cheap as hell. I also set up ST's toolchain which took ages to correctly configure but eventually worked nicely.
>>1247296
Voltage division/drop. I'm not really verse in electrical principles, so I figures I'd ask here
>>1247307
Got it.
>>1247314
There are two 5v pins which are connected to the 5v input.

[Anonymous No.1247355]

>>1247295
>My question: what is the most stable way to connect them? If I connect them parallel wouldn't the 5v divide between both components? >
I know the raspberry pi can power the arduino from its 5v pin, but its the same line as the 5v main rail, so that's essentially parallel as well.

Your house is supplied from a step-down converter (line transformer).
All the outlets in your house are (more or less) in parallel.
They're even in parallel with the lights.
Do you see where this is going?

[Anonymous No.1247429]

>>1246900
There's lots of options for software, they all do mostly the same thing, so doesn't make a huge difference with which one you settle on. I use the SW4STM32 a.k.a System Workbench, it's a single package with toolchain and Eclipse based IDE, works fine for me and all free software. For generating projects and hardware configuration you can use STM32CubeMX.

[Anonymous No.1247439]

>>1247429
>>1247320
>>1246941
Thanks for the tips guys.

I managed to get the IAR workbench IDE working with the STM32Cube stuff, so I'm pretty happy.

I figure by the time I grow out of the size requirement I'll know enough about embedded programming to know what features I need.

[Anonymous No.1247538]

>>1247295
If you're using the Pi3, you'd want to make sure you have at least a 2.1, if not 3A adapter. And actually buy an official one, Pi3 is very sensitive to voltage drips. Mine never crashed, but I've had it with a solid low voltage warning despite a 2.1A adapter that worked fine for everything else.

[Anonymous No.1247539]

Is there something like an electronic compass that I can read out with a uC? I think some of my projects might be cooler by knowing the direction they're facing.

[Anonymous No.1247540]

>>1247539
Nevermind, found one myself.

[Anonymous No.1247633]

I want to make a high frequency (1GHz or so), high power sinusoid as cheaply as possible. Since driving GHz crystals looks like a pain in the arse, I decided that using a crystal resonator would be the way to go, but all the ones I found made sawtooth waves. By filtering this waveform with sequential low-pass filters I could get a signal closer to being a sinusoid, but it was still not perfect. Would I be better off using this to drive a same-frequency crystal oscillator to get a sinusoid out of, or just having a long chain of 5 or 6 low-pass filters with a few RF amplifiers in-between? Getting a 30dBm+ RF amplifier isn't that tough, so the high power part is sorted provided I can get a sinusoid to it. Pic related is a sinusoid generated from a square wave with three filters, and I'd rather have a significantly higher degree of accuracy than this, as the bottoms of the waves are noticeably thinner than the tops.

[Anonymous No.1247634]

>>1247633
So any ideas?

[Anonymous No.1247655]

>>1247633
>linearity
Probably because your output impedance is asymmetrical. Consider a matching transformer.
>cleanness
I presume you're not modulating your signal heavily. Why not a single resonant LC circuit, maybe two if you're really a purist?

[Anonymous No.1247663]

NEW THREAD

>>1247660
>>1247660
>>1247660

NEW THREAD