/diy/ - Do It Yourself

>>1430418
does anyone make a usb dev board that simultaneously handles hid and 1.8v i/o?

>>1430423
i ordered kapton tape but i am not very optimistic at this point
To read a temperature i charge a capacitor through the thermistor and time how long it takes to charge to a certain voltage and then convert it to resistance (since gpio on rpi can't read resistance directly) which i then convert to temperature based on a resistance table for that thermistor. I checked it with multimeter and there is no problem in this part.
So it's either not enough to just press the thermistor against the bed and i need to like sink it into thermal paste or some shit or the thermistor i just a piece of chink shit

>>1430418
>good source of radioactivity edition
kek'd

>>1430406
>Tubes have clips, no soldering.
That's good to know, I guess a simple fuse clip would suffice? What about for the centre electrode?

>Which component has the EMIT OUT pin?
The LM311, since it's a single comparator in a DIP-8 package.

>Keep the capacitive load of the tube as low as possible
I'd drive the audio with a transistor either way, I don't think direct drive would be sufficiently audible.

>>1430431
I'd look up capacitor-ramp ADCs, they were used by old joysticks and are basically what you've described, except more linear since they charge the capacitor with a constant current and so easier to calibrate. There are also digital thermometers in TO-92 packages that might suit your fancy.

I want to have some stuff like radio and light on my garden, but i don't have electricity there.

So i was thinkin i will buy a solar panel and connect it to an old car battery there.
The panel i can afford costs $50 and outputs 17,8V-1,12A, 20W.
I i calculate correctly that means it can fully charge the battery in only 70 hours right?

I want to put a 12V voltage regulator between the panel and the battery to make sure the battery is not too badly raped.

But what is some easy and cheap way to convert 12V battery DC to 230V AC?

>>1430433
Car batteries will not be ideal for your purpose, a deep-cycle battery will be better suited, though I'm sure there's documentation or at least anecdotal evidence for and against the use of car batteries. I don't know the power rating of your battery but it's [Wh = W*h], so you can't really miss it. But make sure you use something along the lines of a purpose-built solar charge controller, not just a generic switching regulator.

To convert to AC you'll need an inverter, but a lot of devices don't actually directly use 230VAC. You could easily run 12V lighting, same for the radio, and USB slots and such can easily be found in models that convert from 12V.

>>1430434
>Car batteries will not be ideal for your purpose,
If i wanted something ideal i would get a gas generator or something, but i have an old car battery so i am be using that since voltage regulator costs pennies and the only big price here will be the solar panel

>>1430429
it might be easier to make your own than to find a board with adjustable Vcc(io). STM32 boards are pretty simple, just copy the bluepill schematic as a base

>>1430431
yeah, round things taped to flat beds don't generally have good thermal contact
there are thermally conductive adhesives. try one. I like STARS-922
also make sure your thermistor is rated for the temperatures that it will see

>>1430432
>using a semiconductor thermometer
>on a heat bed
come on now

>>1430434
>switching regulator.
Why? Solar panels aren't some magical devices, the DC from the is just regular DC.
Normal voltage regulator will work fine, the only thing you need to handle is disconnecting it when the battery charges.

>>1430438
>a purpose-built solar charge controller, not just a generic switching regulator.
>hurr durr yeah let's use an LM317
go back

>>1430436
>it might be easier to make your own than to find a board with adjustable Vcc(io). STM32 boards are pretty simple, just copy the bluepill schematic as a base
thanks anyway, but i'm trying to avoid smt if possible.

>>1430439
>hurr durr let's waste money on overpriced shit
ok

>>1430440
>18V
>b-but no SMT
are you for real right now

>>1430441
show us your muh linear regulator charge controller that doesn't abuse the battery, then

>>1430442
fug
>1.8V
still, I don't know of (m)any 1.8V logic devices in DIP. 1.27mm SOICs aren't that bad, certainly not like 0.4mm BGAs. if you really really do not want to deal with interposer boards, you might have to add some level converters, which are at least potentially available in DIP.

So if i have the voltage regulator, and i want to adjust it for say 6V, i need to put a proper voltage on the adjustment leg.
In a tutorial they do it by connecting a resistor to the input voltage of the regulator and then that resistor to the adjustable leg (pin).
This works fine if your input voltage is stable, but what to do, if it is variable? Because if the input is say 12V and drops to 9V then it will cause the voltage regulator to readjust itself for like 4V or whatever.

The only solution i can think of would be using a separate power source, like a 9V battery, for the adjustable leg, would that work, or does the voltage on the adjustable leg has to be from the same source that is connected to the input on the regulator?

>>1430447

So you're saying you have a voltage regulator that only works if the input voltage is regulated.

Is this your first circuit?

>>1430448
the regulator is adjustable and i need to adjust it to a set value but i can't use the input voltage since its variable

Well i thought i could build the dc ac inverter myself but i see it's kind of complicated so i will just a finished one
I just wonder if there will be any problems if i connect it to the car battery while it's being charged by the panel (through the voltage regulator)

>>1430447
The output voltage is used to feed the voltage divider.

>>1430432
>centre electrode?
Either modified fuse clip or wire coil/spring. Did you test load the 400V with 10MΩ to find regulation and converter efficiency yet?

>>1430451

can you do something like this?

>>1430516
Yeah, that seems like a good idea actually.

I just hope it won't get too hot since i need to step down 20V to 12V

>>1430526
make sure you put this whole thing into an enclosed container
also have an open flame inside so that the ions can move more efficiently

Why does this fucker have 8 pins?
I thought dc/dc convertes have +- in and +- out
that is 4 pins total, not fucking 8
this does not compute

>>1430526
srsly tho, if you want your battery to survive more than two good sunny days, you'd better use a proper CV/CC charging circuit. there are some based on the LM317, even, if you really must go that route

>>1430540
because that's not the whole dc-dc converter, just the controller

>>1430541
>proper CV/CC charging circuit.
But why?
If i make sure the circuit is disconnected before the battery charge is too high or too low, and the regulator makes sure the battery is getting steady 12V, then why exactly would it get damaged in anyway?

>>1430543
>steady 12V
you need to STOP right here and go look up lead-acid battery charging so you know what you're doing. you do not appear to understand the voltage vs. state of charge of a lead-acid battery and you will almost certainly break something

>>1430543
for your perusal
http://batteryuniversity.com/learn/article/charging_the_lead_acid_battery

>>1430545
Eh, fuck it, if i need to get a PhD in fucking battery charging to make the regulator, I will just throw money at it and buy this cunt instead.
That way I can use the extra LM317 to rig up an USB plug and charge my phone with it by connecting it to that load output on the solar regulator.
It's still lots of volts to step down from 12V to USB 5V but since the phone only draws 1A, i should be able to handle the heat by attaching an old cpu heatsink to the LM317.
Altough later i will buy a proper buck converter from Ping Pong so i don't have to waste all that energy as heat.
Hm... i should properly try it on an old phone first just to be sure though.

>>1430551
holy shit ping pong scores again, the local jews are selling bucks for ten bucks a buck and Ping Pong's friendly ali store sells bucks for four bucks a ... LOT a fucking lot, as in 5 pieces for the half price of a single unit, at which they sell them at the local momberg & and popstein store
And people wonder why local economy is in the shitter. Maybe tone down the 1000% overjew on every item and then we can talk about me or anyone shopping locally.

>>1430603

>>1430622
buck converters

>>1430603
running a retail business isn't cheap, sadly. perhaps in the brave new communist world to come, people won't be keeping score so obsessively. i mean, that's kinda how we're getting Allwinner dev boards for $9
btw did you look up actual lead-acid chargers? here's your charge management Ph.D. on a board, for $7 https://www.aliexpress.com/item/MPPT-Solar-Dildo-Controller-3A-For-12V-lead-anus-Battery/32877123788.html

>>1430677
that's a great looking board for the price. too bad the controller is nameless.

>>1430680
there are similar devices for Li+ being sold with a CN3791, some of the sellers are being memey in their shooping. I suspect it's this one
http://www.consonance-elec.com/pdf/datasheet/DSE-CN3767.pdf

>>1430680
>nameless
If you’re good with a DMM and scope you could slap a Fluke or Rigol brand on it.

>>1430689
>Impying Rigol is a brand worth bragging

>>1430514
As is the thing has a 10MΩ load as the divider, and I tested it with the DMM right up to it so that would be 5MΩ, and it could easily get above 450V. I'll add the current shunt to see how the current acts, though I might just test the roasted geiger tube first. After checking for ripple, that is.

>went a little too crazy with the 22µF ceramic caps because didn't have any 2.2µF
>flashrom burner resets every time I clip on to the flash
>mfw

>>1430406
>Keep the capacitive load of the tube as low as possible (pF) or it won't live long.
All the circuits out there I see use a DC-blocking capacitor between the geiger tube and the amplification transistors, is it fine because the transistors don't pull much current? I've also only got a 10nF HV cap, aside from maybe one of my variable caps.

>>1430786
he means on the high voltage side. the energy in that capacitance is being dumped through the tube when the radiation causes it to arc (or whatever the mechanism is, i forget) so it causes heating. nothing on the low side stores any energy so it doesn't matter.

>>1430789
Ah, so it's about having a capacitor in parallel with the tube. Thanks for clearing that up, that makes plenty sense.

So gents, I guess most of you know these 433MHz RF links. I am trying to create a wireless blinky using these. It's pretty simple and all if I do it the convention way of two microcontrollers, one at the receiver end, one at the transmitter. But I am planning on doing it differently.
What if there's only a microcontroller on the transmitter end, and on the receiver end, there's a band pass filter (of a particular band that it allows). Wouldn't the band pass filter allow only the signal from the transmitter to pass assuming they have the same frequency?
I am not a professional, just a hobbyist. I've come to enjoy analogue part of electronics a lot more than expected

>>1430786
>>1430789
I also happen to be making a Geiger counter, but I've seen nothing about using a capacitor anywhere in the circuit. Can one of you elaborate for me?

>>1430841
The 22pF cap here.

>>1430841

>>1430844
My original question was asking about the second cap in that image, not the first. The first cap is the example of what you do not want.

>>1430845
i actually can't think of a reason in my diagram, i was just drawing it from memory. i guess this should work just as well. the other anon's pic is clearly just to decouple the low and high voltage sections but i'm not sure why that topology would be any better. i'd think it would have a slower recovery time.

>>1430844
So I need a capacitor between the center of the output voltage divider and whatever I'm outputting to? In my circuit I just have the output of the voltage divider directly connected to an analog pin on my arduino, the pulses are high enough to measure with no amplification.

>>1430850
>this
if the high side limiting resistor shorted you'd cook your tube and transistor as the hv supply bulk cap(s) discharge but it's not like anyone's making these for mass production.

>>1430853
no i don't think so. i was drawing that thoughtlessly.

>>1430418
I'm trying to make a power supply for my ts100 soldering iron, and I'm confused on how to ground the tip. Should it be connected to earth or just circuit ground?

>>1430853
If you're extracting the signal from the high-side then you need an isolation capacitor, but form low-side you don't. I don't have such a cap lying about so my one will probably be low-side. I think you get a higher signal:noise ratio with high-side signal extraction, at least if you have a larger resistor on the high-side than the low-side of the tube.

I don't suppose there's anything wrong with pic related, since it has a 10M resistor? Though I'm unsure what kind of signal I'd get, I think a small spike?

>>1430857
I've had no problems with noise using the signal from the low side, at least just measuring background radiation.

>>1430858
What tube are you using?

>>1430440
>2018
>afraid of SMT

But more seriously, hard to avoid SMT nowadays. Get some magnification and you'll be OK.

>>1430862
An SBM-20

>>1430856

- connect to circuit ground to avoid any induced voltages leaking from your circuit to the tip.

- connect to earth ground to eliminate static buildup.

- connect to both to get twice the advantages.

personally i dont want my tips connected to anything, coz i will be soldering onto live joints on PCBs and dont wanna short out the power supply if it happens to be grounded.

>>1430677
Can it even be legit mppt for that kind of price? Even the cheapest pwm trash costs twice that.

Shit, batteries are fucking expensive. The car batteries arent that bad, I get 45AH for about $60 but i get barely half that with a deep cycle battery.
I wonder how long will a car battery last if i will be deep cycling it,

>>1430973
>I wonder how long will a car battery last if i will be deep cycling it,

Off the top of my head, they drop to like 60% of their initial capacity after just a few dozen cycles, and it just gets worse from there.

>>1430976
What is the maximum charge i can consume before recharging to not be considered a deep cycle?

>>1430980

"Deep cycle" is just a qualitative term; there is no technical definition for it. I've heard anywhere from 60-90% DOD being called "deep cycle", but I've seen 80% more than others.

It doesn't really matter, in any event. There's no real hard cutoff where you start to see much more significant effect, short of taking it to like 95%+ DOD where you really start to cut into the overall life of the battery. Cycle life just gets progressively lower the further the battery is discharged, with the effect being substantially more pronounced in car batteries with their porous electrodes.

>>1430982
Well i can see cycling to about 50% in a session when we go to the garden
If the battery lasts at least a year it will be enough, spending $60 per year on a new battery wont kill me
It will also be constantly charged by the solar panel

>>1430984
I ordered the panel and now we play the waiting game.
I bet those retards from the post office will manage to crack it.

is the given data of an electric contactor (e.g. switching power/current of http://www.newark.com/schneider-electric/lc1d18p7/iec-contactor/dp/20K1136)) meant per phase or overall? i found contradictory information on the matter so far

What is the easiest way to drive something like pic related (10 leds in a bar-graph pattern) to show progress using an arduino using a minimal amount of pins? I was thinking using an MCP23017 IO-expander and just toggling each led directly, but is there a "smarter" way?

If a capacitator is rated for 50V and it says on it 10u, does that mean that if i connect 25V to it it will charge to 5u max?

>>1431002
10uF is the Capacity value of the capacitor and is more or less fixed (might decrease over the years)

if you connect 25V to the capacitor it will charge to 25V. 50V is the maximum allowed voltage and if you go higher it might/will damage the capacitor

>>1431001

Shift register. If it's one that can sink/source more than a few mA on the outputs (most can), you don't even need any additional hardware other than the obvious current-limiting resistors. A 74HC595 is what I usually reach for.

>>1431004
so even 1V will charge it to 100% capacity but slower?

>>1431002
>>1431004
In addition to what this guy says, charge in a capacitor Q is measured in coulombs, where Q = C*V. This convention isn't used very often (outside of amp-hours), so expect to see people referring to voltage stored in a capacitor. But if you connect two capacitors at different voltages together, it's charge that is conserved, not potential energy (excess is radiated away as heat). The energy in a capacitor can be calculated as E = 0.5*C*V^2. Capacitance is calculated as C = e*A/d, where e is the permittivity of the dielectric, or the relative permittivity multiplied by the permittivity of free space, A is plate area, and d is plate separation/dielectric width, so capacitance is therefore not dependant on voltage.

>>1431008
Capacitance is the measurement of the size of the capacitor, not the measure of what's stored in it. The units are coulombs-per-volt, it tells you how much charge is required to shove into the capacitor for its potential to increase by a single volt, and this ratio is constant regardless of potential.

>>1430837
not very reliably over AM. you'd have to set a pretty long debounce timer and a pretty narrow filter to make sure it was your signal, maybe even a tone sequence. I did this a long time ago with a tone and a tone decoder, transmitting over 70cm FM, and it was good

>>1430965
>From I-V curve of photovoltaic cell, under a given temperature, the photovoltaic cell’s voltages at the maximum power point are nearly constant regardless of the different irradiances. So the maximum power point can be tracked if the photovoltaic cell’s output voltage is regulated to a constant voltage.
>The maximum power point voltage is decided by the following equation: VMPPT =1.205×(1+R3/R4)
it's faux MPPT that depends on a rather presumptive approximation, but it could be better than nothing

>>1431001
depending on other system requirements, you could use:
>the I2C master with an I2C port expander, which would use no additional pins if you're already using the I2C master
>the SPI master with an SPI port expander and an additional chip select, which would use only one additional pin if you're already using the SPI master
>the SPI master with an SPI LED driver like the MAX7219 and an additional chip select, which would use only one additional pin if you're already using the SPI master, and has built-in current limiting and digital brightness control
>a dot/bar-graph driver like the LM3914, fed with an analog voltage via PWM, which would use only one pin and has built in current-limiting, but it would also use one PWM

>>1431008
>>1431002

Capacitors do not work that way.

The capacitance rating, in Farads (or µF/nF/pF/wtF) is a description of how much charge it will take per-volt. It does not charge "up to X.XµF", it charges until it reaches equilibrium with the supply voltage, and how many coulombs that takes depends on its capacitance.

1V will charge it to 1V, or 25V will charge it to 25V; the total energy/charge stored is irrelevant in that respect.

>>1431009
>>1431011
Let me rephrase the question. If i have two identical capacitators, and charge one with 10V and other with 20V, will they both keep a LED on for the same amount of time (assuming the current going through the LED is same in both cases)
?

I've swapped the old and run-down 4.3v 18650 cells on a laptop battery with 4.2v cells I had from another donor. It reports charging them to under 4.2v because they have some use already. Will they burn my house down? I thought overcharging a 4.2v battery to 4.3v was fine but would just give you less cycles.

>>1431012
No they do not. One stores twice the charge (and 4 times the energy) as the other.

>>1431012
no because one capacitor has stored more energy E=0.5*C*U^2 than the other

>>1431014
>>1431015
Thanks, that is pretty interesting, i assumed that doubling the voltage will also only double the energy stored.
So that was why my capacitors kept smoking and blowing up.

>>1431018
They're probably smoking because you're pushing them to too high voltages.

Wait a minute. I just looked at my tiny powebank and it says it has 20000mah and a car battery has 40ah.
Does that mean that two of those cheap ass powebanks have same capacity as a car battery?
Then what the fuck do i need those expensive pieces of shit for?
Looks like the only advantage of the car battery is higher current which i don't need.

>>1431036
no, the capacity that matters is watt-hours
~80Wh < ~480Wh

if load draws only the current it wants then why do i need to know the current to calculate a resistor? for example if a led is rated for 2v i do not need to know what current it needs if i connect it directly to a 2v battery, but if the battery is 5v i suddenly need to know the current it needs in order to calculate a resistor value, why? i only want the resistor to drop voltage across it by 3V, not amperage

>>1431005
another nice item to have is the TPIC6C595, which is like an HC595 with open-drain 100mA MOSFET outputs

>>1431099
>i do not need to know what current it needs if i connect it directly to a 2v battery
oversimplified to the point of wrongness, but we'll let it slide for a sec
>i only want the resistor to drop voltage across it by 3V, not amperage
but voltage and current are interrelated by Ohm's Law. E = I * R. basic algebra, you need to know two to calculate the third. since the variable you can directly control is R, you need to know (available) E and (chosen) I to select R
the relationship of voltage to current through an LED or other silicon diode is non-linear. that red LED will conduct about twice as much current at 2.3V than at 2.0V, and will conduct almost no current at 1.0V. what's more, the LED heats up as current flows through it, and the LED's voltage at a given current drops as it heats up, which makes it want more current at the given voltage, which heats it up more, which, if unrestrained, eventually leads to it gorging itself on current and a very small kaboom. hence, always restrict the current going into an LED

>>1431099
>i only want the resistor to drop voltage across it by 3V, not amperage

figure out what a "three volt" resistor is, and you will answer your own question.

>>1431113
>always restrict the current going into an LED
if you supply exactly 2v though that would be 0V/(20ma/1000) = 0R and that doesn't make sense

>>1431134
see the graph >>1431113 for current-voltage relationship at about 25°C. now subtract about 2mV/°C as the LED die heats up from losses, etc. and then find the new If for that adjusted Vf. more current → more losses → lower Vf → more current and so on. you'd have to find an operating point that's stable under the actual conditions *that* particular LED is in, or actively cool the LED, or otherwise current-limit the supply to the LED, in order to keep current from running away to damaging values. yay engineering
all this is assuming your ideal 2V supply can supply infinite current. in reality, it probably doesn't and probably has some internal series resistance, which is why https://www.instructables.com/id/LED-Throwies/ happens to work

>>1431134
>if you supply exactly 2v though that would be 0V/(20ma/1000) = 0R and that doesn't make sense

Maybe working with electronics isn't in your future.

If i open PC PSU how long after turning it off do i have to wait to touch it without dying?

I have a car from 86 and the radio sounds like ass. Basically everything sounds like am radio quality, but with even more stank, and not the good kind of stank. Would replacing the 32+ year old capacitors fix it? I don't want to replace the unit with a modern one because it looks cool as fuck and I want the car to be as stock as possible.

>>1431267
no promises, but it couldn't hurt matters and it's certainly worth a try

How exactly does the reset pin on a 555 work? How can I use it to 'pause' the circuit? The datasheet is a little ambiguous, at least I don't really get it.

>>1431315
does a schematic help?
looks like it turns on the discharge transistor, resets the flip-flop, and forces the output low
I could be reading it wrong tho

>>1431315
>>1431318

https://en.wikipedia.org/wiki/555_timer_IC

per wikipedia: Reset: a timing interval may be reset by driving this pin to GND, but the timing does not begin again until this pin rises above approximately 0.7 Volts. This pin overrides TRIG (trigger), which overrides THR (threshold). If this pin is not used, it should be connected to VCC to prevent electrical noise causing a reset.

meaning, it overrides everything, as a RESET should.

>>1431315
Pull it low, the 555 timer will stop completely.

Ok I ordered an SMB-20 in case the SMB-21 doesn't work, it was marginally cheaper anyhow. I'd have preferred something better at alpha but those end-window tubes are all like $50 or more, to say nothing of the pancake tubes.

>>1431359

underrated post. 10/10

>>1431364
>underrated post.

you can't do that on the next post.

jeez.

>>1431364
are you being sarcastic because /blogpost

This counter uses a piezo directly in parallel with a resistor in series with the main filtration cap. The maniac is also using zeners for feedback, but that's besides the point. I wonder how loud the clicking is?

Fear

H-bridge ic safety question
If I drive both inputs high (e.g. in1 and in2) on an L298 will it release magic smoke (short Vs to GND), or does the chip have some built-in protection against this?
Skimmed the datasheet and didn't see any mention of having protection against this, but I don't really want to risk frying a motor breakout board just to test it.

Side thought, if it doesn't have any protections against driving both inputs high, could I slap a jelly bean XOR (7485?) before the inputs and make it reasonably safe?

>>1431450
Jesus that's awful.

>>1431451
>Skimmed the datasheet and didn't see any mention of having protection against this, but I don't really want to risk frying a motor breakout board just to test it.

Look at the block diagram on the first page. You can see that In1 turns the upper left transistor on, and, at the same time, turns the lower left transistor off. In2 does the same for the right half.

If one is high and the other is low, you have a current path from Vs, though OUT1/OUT2, and down to Pin 1 (GND). If both are high, neither of the low-side transistors is on, and there is no current path from Vs to GND. Similarly, having both pins low leaves the low-side switches on, and the high-side switches off, with no current path through the bridge.

So, to answer your question directly, nothing will happen.

>>1431459
Oh, I must not have looked to closely at the and gates. Thanks m8

So linear V regulator basically wastes all power it cuts out and a buck converter is actually smart and doesn't used the power it cuts out because it uses a square wave right? Should I use an inductor in series with the buck converter? I worry about voltage spikes in the square wave.

>>1431551
any buck converters you can buy will have an integral inductor.

>>1431551
It's better to think of it in terms of voltage/potential. Your load wants say 3V at 100mA, and you have 12V to give it. You put it through a linear regulator, and to ensure that the load only gets 3V, it has to drop the remaining 9V across itself, and since it also has to have 100mA through it it has to dissipate 900mW of power. On the other hand, an ideal buck regulator will take only 25mA in, convert the 12V to 3V, and output 100mA, dependant on the load of course. Buck converter modules can be treated like a standalone voltage converter, and will have some ripple rejection rating. Meaning your supply voltage and can waver but this will not effect the output voltage very much at all, and the same can be said for variations in the load current.

You can get these modules pretty cheap and you just plug wires into each end and you're good to go, but you may want to get a buck-converter IC and build the circuit surrounding it, which means adding all the filtration capacitors, inductor, diodes, etc. With few exceptions these types of converters are not found as standalone ICs, none that I know of.

>>1431557
I ordered this one
Basically my plan is to use it to charge my various devices, like a tablet, or a phone by using any batteries I have lying around. So I could for example mix a bunch of AAs or 9Vs, and then use this thing to step it down to nice clean safe 5V DC

>>1431572

What are the pros and cons of analog delay lines? They seem like magic and I'm already coming up with ways to compute some crazy shit that wouldn't be possie without them.
>being able to do analog computation with readable/writeable memory
Fucking magic I tell you.

>>1431573
That's just an image from the XL4016's datasheet, which could be pictured in either the purchasing page for an XL4016 or the buck converter module that includes it.

>>1431572
Also I'd use a buck/boost-converter instead of a strictly buck-converter so I could run it off any voltage.

>>1431557
You can get them as modules that a ready to be placed on circuit board.

>With few exceptions these types of converters are not found as standalone ICs, none that I know of.

>>1431573
It's this thing
https://www.aliexpress.com/item/MCIGICM-Original-DC-DC-XH-M401-buck-module-XL4016E1-high-power-DC-voltage-regulator-with-maximum/32748543183.html?spm=a2g0s.9042311.0.0.42c74c4dlnTwDh

Looks pretty robust

>>1431589
In my situation I will really never be In a case where a voltage i have is lower than the 5V.
And if i am in such a situation, then it will be probably like i have a single AAA 2.5V cell and if pump it up to 5V then the current will be such a joke that the phone will be like "is--is it in yet?" and won't charge for shit

>>1431602
A single AA is 1.2V, not 2.5V.

>>1431134
If you just apply the LED's nominal forward voltage without any current limit, the result will be anywhere between no light and a dead LED.

Above the forward voltage, small changes in voltage produce large changes in current. The actual forward voltage varies with process variation (different LEDs even from the same batch will have slightly different voltages) and temperature. Most power supplies (batteries or PSUs) aren't accurate to a millivolt either.

To get reliable behaviour, the supply voltage needs to be at least 25% more than the nominal forward voltage, so that small changes in either voltage don't produce large changes in current.

The other option is to use a constant-current regulator (either linear or switch-mode), but that's a lot of extra complexity that's best avoided (and a constant-current regulator will also have a voltage drop).

>>1431603
1.2V for rechargeable, 1.5V for primary cells (e.g. alkaline).

>>1431603
>A single AA is 1.2V, not 2.5V.
FUCK looks like i dimension hopped again, fucking Bernstains all over again

>>1431606
While 1.5V is the nominal voltage for an alkaline cell, 1.2V is a more useful guess for actual use. And 800mV per cell is a typical cut-off, if you want to get most of the juice.

>>1431013
Did test them all under load? Wat was the lowest/highest voltage readings? Laptop bms is prolly charging to lowest spec. Doubt if they are cause for concern, though. If worried, just test every six months.

>>1431267
>Remove radio
>Strip PCB's of everything other than switches, knobs, screens, and the tape deck
>Rebuild with new controller and decent shit
What I'm doing to my car's head unit so I can use it's screen to do whatever I want.

Question time:
I'm building a solar glider designed to fly infinitely in the stratopause, but that shit's cold.
The problem is that above and below this slim range, shit's even colder.
I doubt the solar cells would mind traversing this range all that much, but trying to find a suitable battery technology is proving difficult.
I am thinking lithium cells submerged in Isopropyl with nichrome wires for heating, all in an insulated tube, but I'd really rather only have to heat this all with an outside temperature of 0c, not -60c (-80f+/-).

All I need is a technology suitable as a backup battery, to do nothing other than power the flight controller and servos 'till the plane can thermally soar back to a suitable height.
I'd happily burn through the lithium bank's charge to do nothing more than keep themselves warm, I'd be flying well above the clouds so energy from the plane's panels will always be more than enough, but overnight, should the cells get too low, the batteries depleted, I'd need something to keep it going till morning because, if it dipped below the cloud line, it'd be game over for the plane.

>>1431647
I'll be honest, if engineering teams from the likes of Facebook and Google have tried making perpetually flying drones, and mostly failed, I'm not sure an anon from /diy/ could do it.

See:
https://www.popularmechanics.com/flight/drones/a24716/alphabet-kill-titan-aerospace-drone/

https://techcrunch.com/2018/06/26/facebook-permanently-grounds-its-aquila-solar-powered-internet-plane/

good luck though

>>1431643
excellent choice, anon

>>1431647
are you sure you have enough pressure to keep the IPA liquid? insulation and a double-walled pressure vessel might be your best bet, and it's lightweight

Shit, seems like my solar charge controller refuses to put any voltage apart from 5mv onto the load input without battery present

>>1431667
try a large electrolytic capacitor with a few ohms of power resistor in series

>>1431667
Also the manual says it is intended for open, agm and gel batteries. My cars battery is flooded, will it work with it?

>>1431668
I think the problem is that the controler uses the battery to power itself so with just the panel connected it remains off

>>1431669
flooded is probably the same as open, even if maintenance-free

>>1431670
a large electrolytic with a resistor in series is a very common mock battery arrangement, used and recommended in many datasheets for testing. of course, if you have a battery to hand, use it

>>1431671
Thanks, can, you give me specific numbers for the cap and the resistor?

>>1431675
idunno, 1000µF and 10 ohms?

>>1431652
I've thoroughly gone through all of their attempts, along with attempts from independent groups, and most have either done something stupid, or had artificial restraints.
The current record holder for the category and size of craft I'm building, AtlantikSolar, was utterly restrained by a pitiful height limit, and on top of that, had no soaring algorithms.

Soaring is far more important than anything else here, microsoft did wonderfully in their own experiments. The only reason I'm bothering with solar at all is so I can keep driving the servos perpetually to steer, and, when the batteries are fully charged and warmed, and the panels are delivering surplus energy, burn only what's otherwise useless to drive the plane a little faster (soaring can be bloody slow).

On top of that, AtlantikSolar, due to the height restrictions, could not use height as a way of storing energy.
I, however, am going to go as high as a planes wings can possibly function, and, during the night, even if there aren't any decent thermals to catch, use the energy stored in the height alone as a form of propulsion.

With either of these I should be able to achieve what I want, let alone both at once.
And I don't actually want to fly perpetually, just long enough to cross from Australia to California, circle the globe, and break every aircraft flight time distance world record there is. Simple.

>>1431658
It'll be insulation inside a pressure vessel, don't worry. Keeping the whole thing from exploding at the low air pressures is actually one of my main concerns, and my stupid understanding of science tells me if the vessel is sealed without gasses everything will be fine.
I won't try to fly the thing until I have tested the electronics extensively in vacuums, dunked them in liquid nitrogen, and pressurised the inside of the chamber as well until it bursts, so I know what my limits are.


>>1431667
Is it designed to only function with a battery present?

>>1431677
Ok i have 2x 560 uf from an old psu, i will put then in paralel and precharge them to 13v with an ordinary lithium battery and then connect them to the charge controler so it can boot itself from them and start working, will that work?

>>1431684
>will that work?

possibly but not very likely. 1100uF doesnt hold a lot of energy, only enough to light an LED for a couple of seconds. i'd start with 100x that much.

>>1431751
I'll have to go to the electronics store since i don't have anything bigger than the PSU caps. I'll take the biggest one they have. Maybe if i'm lucky they will have one of them supercaps that have like 2F

>>1431759

supercaps tend to have a top voltage of around 2.5V so you'd need a bunch. some are designed for 5V operation, but you'd still need 3.

there are youtube videos where car dudes have replaced their battery with a series of supercaps. might be worthwhile to see how they've wired them. e.g. do they use balancing resistors?

>>1431759

supercaps tend to have a top voltage of around 2.5V so you'd need a bunch. some are designed for 5V operation, but you'd still need 3.

there are youtube videos where car dudes have replaced their battery with a series of supercaps. might be worthwhile to see how they've wired them. e.g. do they use balancing resistors?

if the cost of that is too much, might wanna try a 12V gel cell, as used in house alarms. or even a used car battery. a bunch of dingy stores not far from where i live sell 'em for $20 with a 30-day guarantee.

>>1431751
it only needs to simulate enough to convince the charger there's a battery attached. a functional simulation is far from necessary

>>1431684
try it
>supercaps
probably not in a 13V rating

>>1431780
Yeah I am thinking of just buying a new car battery. Used ones could be half dead and i would get not warranty here. New one costs about $70 for 45AH and it will come in handy if i need energy when it's raining or i want want to have like a security system running in the night

>>1431781
>it only needs to simulate enough to convince the charger there's a battery attached.

the suggestion was that the battery is used to power the battery charging circuit, which makes for a weak design, but makes sense economically.

>>1431784
it makes a little sense for it to not start up with a battery fault. I'm thinking it just wants to see enough voltage there to look like a healthy battery, then it will un-choke itself out of trickle mode
internal photos would be most informative in this case, ofc

Are 50/60Hz mains transformers compatible with the 400Hz power systems on stuff like boats and airplanes?

>>1431789

steel is good up to about 1000Hz. it's not ideal but should work fine.
and switching power supplies dont care about frequancy.

>>1431679
Better be looking at those paper-thin organic solar panels, friendo.

>>1430845
It's a DC blockign cap. The way these tubes work is that you creat a strong electric field that will break down and conduct when an ionizing particle enters the geiger tube.

The cap is to keep the HV off the transistor base and only to conduct the pulse of the GM-tube break downs.

Each pulse corresponds to an ionizing event.

>>1431318
>responds to someone who don't know how a 555 timer works with a transistor diagram

What did he mean by this?

>>1432016
I understand that, just the guy said to "keep the capacitive load of the tube as low as possible". He didn't explain why nor what capacitor positioning he meant, so I asked in this thread. It was explained that when an ionisation event occurs, if there's a lot of stored charge in a capacitor in parallel with the tube that prevents the voltage from dropping, this will form an arc within the tube, thus shortening its lifetime. Since I'm not using a capacitor in parallel with the tube this is not an issue, but it's a useful thing to be aware of anyhow. But I knew I needed a DC blocking capacitor in the first place, hence why the vague statement prompted me to ask the question.

The electric field within the geiger tube will be strongest when at steady-state, when there's no ionisation event at all, since this is when there will be the full ~400V across the tube. But the dielectric strength of the tube decreases in the case of an ionisation event, and if this avalanche does not cause a great enough drop in electric field (the purpose of the massive "low-pass" resistor) then it is possible that an arc will form within the tube.

Ok my ion chamber circuits make a noise at all times EXCEPT when there's a current flowing between the electrodes. I'm throwing in the towel for today.

I found some local people selling car batteries for cheap.
How can I check the health of those batteries with a multimeter? Or some other way? I can't obviously take them aprt.

>>1432121
>Or some other way?
eBay sells them for <$20

>>1432124
ALSO: Google "make a battery load tester"

>>1432125
>load tester
Isn't that literally multimeter in series with a resistor connected to the battery? I can do that

>>1432126
In parallel.
And just a volt meter.

>>1432127
Current has to be measured in series though, if i want to measure voltage, i can do that with parallel but i dont need resistor for that
i am confused now

>>1432128
>Current has to be measured
If you know the load in ohms you can determine the amps by the volts.

>>1432130
if all i need to do is get a current reading from the battery, thats no problem, but what values should i be looking for in a healthy battery?

>>1432126
>>1432128
quit showing your retardation
just follow advice given >>1432125

>>1432132
Bullying is NOT ok. Stop.

>>1431920
That shit's weak and I don't trust it at -60c.
Raw solar cells, even the cheap-as-chips 4w 6x6" ones, weigh absolute fuckall and are well under 1mm thick.
A stack of god-knows how many of these are still under an inch thick.
Inside a rigid wing I doubt they'd have any problems whatsoever.
(I'm not thinking of putting these on the outside of a styrofoam wing or anything, I'm going to build a skeletal wing, run these along the inside, and encase the whole thing in a rigid film

>>1432017
HE GOD LURN THE HARD WAY, LIEK I DID BACK IN '75
FUCK TUTORIALS, I GOT GIVEN A 6502 WITH A 300 PAGE SPEC SHEET AND TOLD TO BUILD A COMPUTER, WELL, I DAMN WELL BUILT A COMPUTER!

>>1432128
Resistor is used to present a significant load, as rotten batteries have difficulties supplying heavy loads.
Parallel connection with a volt meter is more convenient, because 1) it gives a more direct indication of the battery condition, 2) measuring high currents is more expensive than measuring low voltages and 3) you need to worry less about the wiring and meter resistances.

That said, a resistor in series with an ammeter would work.

>>1432131
>what values should i be looking for in a healthy battery?

you put a bunch of 12V car lights or 12V halogen bulbs on the battery until you get the expected load current, or maybe twice that, and keep track of voltage. when it gets to about 11V, you know you've reached the end of the battery's useful life, so note how long that took. make a graph of time vs voltage. that gives you the battery's fundamental load-handling characteristic. it's like looking into it's very soul.

>>1432152
I can't really do that, I basically drive out to a guy buy the battery and then if it breaks i am fucked. All i can do is like do some quick 30 second check with the multimeter befor handing over the money

>>1432154

ah, then you can only test if the voltage doesnt drop too much under a heavy load, and if it recovers quickly. i suppose a welding rod or 2 will provide a nice heavy load for that kind of quick test.

>>1432155

Welding with jumper cables and a pair of batteries
https://www.youtube.com/watch?v=PV5oLPLUzrM

>>1432156
If you happen to carry around some 2-awg wire, sure.

>>1432155
>>1432156
>"What the fuck are you doing?!"
>"Kek did you seriously think I will buy a battery from you if i can't weld metal with it?"

>>1432156
That can't be good for the batteries, They are designed to only provide very short burst of current upon engine ignition

>>1432167
It's marginally better than dazzling the guy with 500 spotlights. Unless you arc flash him I guess.

So I found out the solar charge controller I have is the shitty PWM.

This means that with a 21V panel it will waste 9V since the square wave will be 0 for like 40% of the time. (In order to lower the voltage for the accumulator charging)
So I was thinking, what if I connected a buck converter to the panel wires and then connected the buck converter's output to the solar charger controller input slot for the solar panel?
Since buck converter uses an inductor to lower the voltage, it increases current and doesn't waste any of that precious precious power. And because of that, the solar charger could keep the pulse with at pretty much 100% on and waste almost nothing.

Would that work?

>>1432180
So lemme get this right
You got 21v 1A in, you get 12v 1A out?

>>1432184
The solar panel has two poles comming out + and -, and provides 21V 1.5A (can be less if there are clouds and such)
Normally the panel is connected directly to the solar charge controller, which pumps it directly to the lead accumulator connected also to it (to the controller). But first it uses pwm to bring to the solar panel voltage to a voltage the battery can handle, usually 12V. This wastes lots of power because of the way pwm works.

So I want to put a buck converter between the panel and the controller, so that 12V is going into the controller and it doesn't have to waste power by bringing the voltage down with PWM

>>1432184
>You got 21v 1A in, you get 12v 1A out?
Also yes, this is how the PWM controller works. There are fancier ones that lower voltage and increase current (as per ohms law) but those are way too expesnive and i already have this pwm shitty one

>>1432187
I assumed any decent charge controller had its own buck converter built-in, I'd just get a good one. Though you might be able to go full Class-D on the PWM output with some low-pass LC filters and get that high efficiency.

>>1432191
>low-pass LC filters
I am not really advanced enough to do that

But i already own a buck converter and the pwm charger, so I want to know If it will work or not the way i described it?
It would be great if yes, since it would get like 30% more free energy for $0 spent

>>1432076
>not using a capacitor in parallel with the tube
the point is to be careful about your connection routing and not create one

>>1432134
hello nu-/b/

>>1432180
you're sure there's no inductor inside to store current?
>solar charger could keep the pulse with at pretty much 100%
but your buck converter cannot. maybe if you add a big cap
but always add a diode to prevent back-flow through the panel. high-current Schottky diodes are pretty easy to get

>>1432141
>not building a 6502 computer from a datasheet back in '75
filthy casual
anyway this is an image board. I used the image to trace out what happens when RESET gets pulled low and OP could have too
>mods asleep, post sinks

>>1432199
>but your buck converter cannot
Why wouldn't it? At the output you will get steady voltage and current. Power loss in the buck converter will be no more than 7%

>>1432180
Solar converters are usually boost-buck or boost-invert. A boost converter draws a constant current, which is typically tuned to match the solar panel's output via maximum power point tracking (MMPT). The output from the boost converter either charges a battery via a buck converter or generates mains via an inverter.

The buck converter or inverter limits its output current to maintain its input voltage; if the input voltage starts to drop, it means it's drawing more power than the panel can supply.

>>1432203
Do you mean mppt? Like i said above my converter is pwm not mppt because mppts are too expensive

>>1432210
All DC-DC converters use PWM. But they normally have an inductor to smooth out the current. A boost converter has an inductor on the input, so you get continuous current draw but pulses of output current. A buck converter has an inductor on the output, so you get pulses of current draw but continuous output current.

If MPPT is used, it just adjusts the duty cycle in the converter to maximise power transfer.

>>1432225
Nobody uses simple inductor based SMPSes anymore. That's just for learning the basic theory. It's all transformer based these days; flyback, forward, and resonant LLC converters are where it's at.

>>1432230
>nobody
>there aren't a handful of simple inductor based SMPSes in every device
>everyone wants to pay more for transformers because reasons
sure kid

>>1432237
Pick up literally any piece of consumer electronics that has an SMPS powering it and 9/10 times you're gonna find a transformer in it, or in the case of the forward converter a transformer and inductor.

>he thinks people pay more for xformers
kek, you don't buy off the shelf premade ones. You wind your own. The ferrite cores are generally pretty cheap.

>>1432244
ohhh, are you distinguishing between off-line SMPS and dc-dc converters? obviously, systems connected to the mains and having external connections will be usiing transformers for isolation and they may even throw some extra windings on for multiple outputs because it's cheap enough. yet a large enough portion of devices still keeps mains isolation and ac-dc conversion offboard and regulates at the point of load onboard

welp turns out a car battery can only be discharged by about 20% before it starts dying. that fucking sucks.
deep cycle batteries are so fucking expensive

>>1432292
AGM 12Ah for $50 that is the best i can get. Guess i'm gona go for it, getting 45Ah starter battery for that same price would be pointless if i can only discharge by 20%, which would only give me 9Ah at 5x size and weight

YES it's working, delicious 14v ready for load

fuck you electric companies

>>1432313
So at 14v i measured 0.29A that means this shit needs to eat about 4.19W of heat right? It's rated for 10w yet it gets so hot i cant even touch it, what gives?

>>1432318
they can stand much more heat than your fingers can

>>1432318
So between panel and controller i measure 0.82A so this pwm cunt is robbing of over 30% of the sweet sweet sun juice just as i presumed

>>1432324
Can't wait for the buck converter to arrive so i can test if it will help me salvage at least 20% more

>>1432330
there are versions of those buck converters that are designed specifically for battery charging that will sort of follow the CV/CC profile Pb-acid batteries like. too bad you didn't get one of those for $1.36

>>1432334
I don't need that because the charger handles the special snowflake 3 phase lead charging plus supports switching between them

>>1432313
Interesting, after the controller turns on i can remove the battery and it works fine, this means that i could use it with a large cap instead of a battery too as the other anon suggested

Would this work?

>>1432340
if you don't need much coverage for cloudy moments, you could indeed

>>1432360
*shrug* prolly
you'll want to run the Pi itself from the always-on 12V and use the ignition lead for everything else. you'll also want to sense the voltage on the ignition lead (through a voltage divider of course) and run a shutdown when the ignition lead loses power

Currently working on my little variable speed sign project. Progressing little by little.

Still unsure about what I should use for switching the LED segments.
Currently thinking to use 10x 2N3904 transistors for each string and parallel interface between the LED board and the controller board. Serial interface would require dual 8-bit shift registers and the board might be a bit too small for those. Also then the LED board would need to have a 5 V supply for the register ICs.

The controller device will be located on a separate board. I want to use 4000-series discrete logic ICs for controlling the display. But if it doesn't work out I can also implement the logic with an AVR IC.

>>1432366
looks like there's room enough to squeeze two CD4014s on there (A07-D14, A16-D23) if you don't mount those other screw terminals, and still have enough room left over for driver transistors and two resistors each

Hello guys,
I guess i need a little bit help.
I have an omni wheel platform which we had to programm in uni (all done, but improvments needed). They have three 12V dc motors with encoders on them. Two of them are dead though. Now I searched through the internet, but couldnt find any replacement.
But since it seems to be a rather simple circuit I might just design one myself and fabricate it in my uni.
The only thing I am wondering is what might be the circuit i circled in the picture? Should be a mosfet or something, right? How should the pinout of that be?

PS:
if anyone is interested this is our platform
http://www.nexusrobot.com/product/3wd-48mm-omni-wheel-robot-platform-chassiswith-encoderblack-15001b.html

>>1432387
Here's a picture from the other side. It is just some light barrier (two of them obviously).

>>1432388
Stop putting wheels on metal triangles and build a hot girl robot instead. That is literally the only reason why you wen't into robotics uni anyway.

>>1432387
dual complementary MOSFET maybe. did you look up the code printed on it?

>>1432392
Yes and I only find the dc motors itself with it.
I even found (((the same))) ones from multiple companies. I emailed one of them if they sell only the encoders. Will see.
It would only be 30-40 buckos for three complete new motors, but that's no challenge

>>1432393
what are the codes printed on the SOT23-5s?

>>1432395
>SOT23-5s
Oh nice. Prolly I didn't look close enough and thought they have no code. It looks like it is c145 (maybe 146, or 148, but should be 145)

>>1432334
give a link to that board please

>>1432402
could be 74LVC1G14. those are just on the encoder? they don't drive the motor itself?

>>1432408
https://www.aliexpress.com/item/bunghole-dropper-small-size-pig-sheep/2055097301.html

>>1432409
As you can see in this >>1432387 picture the powerlines of the motor are soldered onto the pcb but i can't see any traces going towards those ICs. So I don't think they are connected to the motor itself.
PS: Thank you based canon. On one its C145

>>1432409
>>1432413
and on another it's c14k. Which seem to be the same then

>>1432411
>Lithium Battery Charger Converter Module
I thought it was specifically for lead/acid batteries.
I already have a couple of boards similar to that.
I'll still keep it in mind.
Thanks.

>>1432413
and pin 1 is no-connect on both? that confirms it. now you just have to be sure it's those inverters that are the problem, and not something with the wheel or the LED/photodiode, or the motor/driver

>>1432415
as a matter of fact, Li+ and Pb-acid are charged with the same general algorithm

>>1432417
>Li+ and Pb-acid are charged with the same general algorithm
I was hoping it had the continue charging to 'over-charge' condition then drop back to maintenance level.

>>1432417
>and pin 1 is no-connect on both? that confirms it. now you just have to be sure it's those inverters that are the problem, and not something with the wheel or the LED/photodiode, or the motor/driver
Nice, thank you! This is super exciting.
Just tested LEDs and Photodiodes and they all seem to behave properly.
I had the working motor as reference and the leds had a voltage of 1.1V over them and the photodiodes had around 0.2-0.4V over them when open and 4.9somethingV when closed.
All of the non-working modules behaved the same. So it's prolly only the ic's. Then I guess i will order them.

Which, seems to be reasonable since on of those two dead ones was damaged by me by connecting gnd and vcc the wrong way and then the magic smoke got released. The other one was dead from the beginning. Great! Thanks!!

>>1432423
mmmmyep, that'll do it. cheers

>>1432230
> Nobody uses simple inductor based SMPSes anymore.
You're confusing a PSU (something which generates low-voltage DC from mains) with a DC-DC converter.

The choice between an inductor-based topology (buck, boost, Cuk, SEPIC, etc) and a transformer-based topology (flyback, half-bridge, full-bridge) is usually dictated by whether you need isolation. PSUs need isolation and so invariably use transformer-based topologies (although PSUs with active PFC put a boost converter before the flyback/bridge circuit). Most DC-DC converters use inductor-based topologies.

E.g. the (self-contained) PSU in a PC will be an isolating topology (usually flyback for anything less than ~1kW), but there's also multiple buck converters on the motherboard for generating 3.3V, 1.8V or whatever from the 12V and 5V the PSU puts out.

>>1432318
This is a trap for young players.

Just because a load is rated for 10Watts doesn't mean you can just dump 10W through it. You need to read the datasheet and calculate how much the temperature will rise based on the power dissipation.

If you look at pic related for example this 10W resistor will rise 200C at 100% load, it just so happens to be rated for maximum 225C which is that 200C + 25C ambient. So you can dump the max through it, but it will be hot as fuck.

In this case you are OK but you might run into shit in the future where the load is rated for 100W+ but they don't explicitly tell you that you must have a near perfect thermal interface to a near perfect heatsink or the part will smoke.

Is there a way to permenantly connect a usb device to a usb port, like soldering or using screws or something?

Is the COM terminal optional?

>>1432613
probably not, if its active it needs voltage to work.

>>1432613
Definitely not.

AAAAAAAAAAAAAAAA

>>1432656
How did THAT happen? Was it bad etching or did something destroy the existing traces?

>>1432660
A battery (encased in blue plastic near the bottom) went bad and corroded the traces

Quick question:

I'm making a 3 wheel cart that's going to be ran by 2 brushless motors. It's gotta carry about 400 pounds (weight of person + load + weight of cart), but it only needs a top speed of like 10mph. How do I go about choosing a motor size and how much would a good pair of motors + esc's cost?

Also posting in QTDDTOT.

I work at a place where we cut thing with a CNC laser. I have always wondered if I could cut the layers of a PCB with the laser. I am fairly certain we could.

My questions is, is there a relatively available board out there that I could download for kiCAD or Eagle that would be easy to source parts for?

Ideally, I was hoping that there would be a kit somewhere for all the non-PCB components for something (Arduino? HAM?). That way I wouldn't have to source a lot of little fiddly bits.

Any ideas where to get started? I just want to download a file and buy the components in a kit.

>>1432581
Sometimes you'll find that the only way to achieve the stated power is to get the ambient temperature down to -200C.

"Lies, damn lies and datasheets".

>>1432662
Ah, I guess if it's using 7400s it's got to be a fairly old battery.

Is there any difference between the drain and source of a real J-fet? I know the concept of a JFET allows current to travel both ways, but are they practically build with asymmetric depletion region or something?

>>1432613

the dashed line certainly suggests it's optional. in the same way that light dimmers only need 2 wires: in and out.

>>1432731
Is that a fucking variable resistor?

>>1432735

it's just a bunch of electronic components. they dont fuck at all coz they're immobile.

>>1432737
Doesn't that mean it's impossible to run those at a 100% duty cycle? Unless it can power itself off the voltage dropped across the TRIAC, that is.

>>1432669

the main reason to buy a kit is so that you dont have to make a PCB using stinky chemicals. and any for-profit kit maker is not gonna wanna give up their design so a competitor could copy it easily. so, it's possible you might find such a thing, but not likely.

the most likely place to find such a thing, i'd guess, is something like instructables.com, where some loser will publish his full project then some company comes along and offers a kit if it's popular.

>>1432742
>impossible to run those at a 100% duty cycle?

97% should be enough for any man, lest he be a glutton bound for hell.

I am confused. Am i reading this wrong?
It says that discharge is 1.75A for 2Hours

But that makes no sense. The battery is 17AH and that means the discharge should be 8.5 for 2hours

>>1432765

That datasheet is some jank Chinglish, but, near as I can tell, what that spec is saying is that getting more than 300 cycles out of it will require that you limit your usage to less than 50% depth of discharge and don't exceed 1.75A.

>>1432769

Note, however, that I also have no idea what percent of the original capacity that they're claiming as end-of-life, so...that seems to be a fairly useless specification all-around.

>>1432769
Wow a 3 kilogram lead battery than can't even charge my phone at full speed if i want it to last
chink technology at its peak

>>1432746
Personally I was more worried about the switching noise than the actual power duty cycle.

>>1432773

since we're at a very low frequency, there's not much noise to worry about. and commercial units usually incorporate a coil and a cap/resistor combo to clean things up.

this theoretical pic shows very straight lines which could mean a lot of harmonics, but real life waveforms are a lot smoother than this.

I need ADB to usb adapter, but I am a broke brainless fuck, what do

>>1432769
>what that spec is saying is that getting more than 300 cycles out of it will require that you limit your usage to less than 50% depth of discharge and don't exceed 1.75A.

no, that's not what it's saying. it's simply letting you know the test conditions they used to get that count of 300.
so, it's no talking about any kind of limits; only how it acts in a typical scenario.

>>1432776
I guess it's hardly difficult to whack on a common-mode suppression choke, some X and Y caps, and the like.

>>1432784
>common-mode suppression choke, some X and Y caps

actually those are impossible since you only have one side of the power line to work on.

>>1432601
yes
(it's metal, just fucking weld it, or pull off the connectors and solder it to the board)

>>1432715
yes
(they are used in special applications where lowest gate-drain capacitance is required e.g. RF)

>>1432789
maybe that's what the common terminal is for

>>1432789
Putting them across each input to the dimmer circuit would eliminate any noise produced by that circuit.

>>1432813
>solder it to the board
This is what I'd do, provided it's easily done. But there aren't any cases I can think of where a permanent USB device is more useful than a temporary one. By soldering directly on you lose some shielding and makes it less optimal for transmitting data quickly, such as to an external drive. I have done this however when experimenting with a portable raspberry-pi SDR receiver; the dongle would protrude too much so I soldered a 4-wire cable onto both the contacts at the back of the female port and directly onto the dongle's PCB (after removing the dongle's USB jack).

What is the point of shortcircuting the output of an opamp to one of its input legs?

>>1432816
I assume you're referring to an op-amp in the buffer configuration. For starters, the term short-circuit implies a large amount of current is being shorted, but the input pins of an op-amp are both very high impedance, ideally infinite. The equation for the output of an op-amp is as follows:
>V_out = (V_+ - V_-)*A
Where A is the open-loop gain (also theoretically infinite). So by shorting the inverting input V_- to the output V_out, we're equating the two:
>V_out = V_-
Or in other words:
>V_out = (V_+ - V_out)*A
We can solve for V_out:
>(1+A)*V_out = A*V_+
>(1/A + 1)*V_out = V_+
Since 1/A is much much smaller than 1, we can approximate this as:
>V_out = V_+
So the voltage at the output will mirror the voltage at the input. This might seem pointless, but if you have a signal coming out of a source with very high output impedance, such as a microphone or phototransistor, putting a low-resistance load on such a signal would try to pull so much current that the voltage would droop. By instead feeding it into something with very large input impedance, you can retain this signal's amplitude and throw it into any manner of low-input impedance circuits.

This same feature is used in non-inverting amplifiers, whereby the signal is fed solely into the non-inverting input, conversely to the inverting amplifier topology. Calculating the V_out from a variety of op-amp circuits can give you a good understanding of how they work, particularly when you start looking at frequency dependancies and transfer functions.

>>1432816
https://www.youtube.com/watch?v=_o4ScgRZtNI

>>1432816
>>1432824
In addition to this, deriving the effect of gain fluctuations and poor frequency responses on an amplifier with and without negative feedback is also very educational. You start with the equation:
>V_out = A*(V_in + B*V_out)
Where A is the open-loop gain and B is your feedback fraction, corresponding to the following circuit:

Holy shit I am so happy I got this working

>>1432852

>>1432852
I hope none of those jumpers are high-frequency lines, though I guess it's only 20MS/s

>>1432854
>20Ms/sec

damn that things better than my 10ms/sec hp from 1980

>>1430418
Question that maybe one of you guys can help me with:

When I'm working on things like restoring a power tool or maintaining my car, or even just using a multimeter, black is common, red is power. Just about everything I can think of is always coded as black = negative and white/red = positive. Why is it that when I'm replacing an outlet or light switch that black is live and white is common?

I'm confused as fuck about this and whether I'm watching Bob Vila telling me that black is positive and white is negative, or I'm reading it in an online article

>https://www.angieslist.com/articles/what-do-electrical-wire-color-codes-mean.htm

I don't get why it isn't standard. If I'm working on a power tool from the 1950's, I look at the cord and find a black common, white live, and green ground. If I'm working on my car, I have a black common connected to the frame of the car, and a red live. If I'm working on a fucking wire connecting a 9v battery to my garage door opener panel, I have a black common, and a red live connected to the little terminal pad. If I'm working on an RC car, same deal, black common, red or white live. Why the fuck is it that I then go into my house to replace a light socket and I have to then think backwards and wire black live, and white neutral when everything else including my fucking multimeter uses the other standard of color coding?

>>1432855
>87
bretty gud, anon

>>1432870
different worlds, different standards

>>1432870

It's frustrating at first, but rather than getting philosophical or questioning the historical reasons, it's simpler to just accept it. I agree with you 100%, but it's not the kind of thing to fret about.

It could be worse, if you somehow had a job or a lifestyle where you did electrical work in the usa for a month or two, then in europe, then japan, etc.

Gonna turn this old psu into a bench powersupply for my tinkering, wish me luck

How do i make it so that my circuit won't supply more than say .1A of current? You know, in case i fuck stuff up

>>1432904
what circuit, dummy

>>1432905
arbitrary circuit

the load connected doesn't matter, i simply want it to be unable to draw more than 0.1A from the power source

>>1432906
try these
https://www.mouser.com/Search/Refine.aspx?Keyword=polyfuse

>>1432906
Use a reallllllllly small wire

>>1432909
I don't want to stop the current flow but to limit it. So even if i say short circuit the thing, it will simply keep sending 0.1A through it

>>1432913
R1 = 12 ohms, 1/4W

>>1432916
So, the R1 adjusts max current allowed? Thats pretty nice, i hope they sell bigger ones too, i need up to 10A in some cases

>>1432916
also, what is the cap for?

>>1432932
it's a voltage regulator. the input decoupling cap helps stabilize the regulator. in this configuration, it regulates so that there's 1.2V across R1. per Ohm's Law, I = E / R
>10A
this is why "arbitrary circuit" isn't an answer
you're gonna need a bigger boat

>>1432669
You want a kit without the PCB but with components? Just buy a kit from any hobbyist electronics store and throw away the pcb. There are plenty of hobbyist stores that publish gerbers and schematics for their kits. They can be a few dollars. How many are you planning to buy?

>>1432669
you can't directly cut or engrave copper clad with a laser, it's too reflective. you'll just reflect it straight back into the lens and break it. what you can do though is use it to cut away an etch resist, and then etch the board like normal.

Would this work?
basically an op amp, on one terminal there would be a voltage regulated by a trimmer
on other where would be a voltage from a parallel connected load.
On the output of the trimmer would be a mosfet and if the load voltage was lower than the trimmer one then the opamp would try to like do negative voltage which wouldn't do shit, but if it was higher, it would do positive voltage and trigger the mosfet that would open a relay and shut off the circuit

>>1432824
So an op-amp, and to a lesser extent transistors can transform impedances like a transformer can?

>>1433042
this is an image board. post sketch

>>1433053
>this is an image board.
That is why I painted a picture with my words
The text form is my canvas

>>1433058
>That is why I painted a picture with my words

I take it you are a picasso fan.

draw a circuit, pablo.

>>1432870
>If I'm working on a power tool from the 1950's, I look at the cord and find a black common, white live, and green ground.
It shouldn't be wired that way. Even in the 50's black was Line (hot) and white was Neutral (ground).
All those black/common and red/hot things you're talking about are DC.
Black/hot and red/hot and white/neutral (ground/return) are AC
Those colors are standard for the type of service in burgerland.
If you are that easily confused perhaps messing with electrical stuff isn't really for you.

>>1433066
It's very simple, I set a desired voltage with the trimmer and then if the load exceeds that voltage, the op amp will saturate the gate on the mossfet which will allow current flow and this will close the relay (it's a special relay which remains open until manually closed) and it disconnects the load from the source.
I attached a schema

>>1432852
You do know that it's acceptable to make jumpers 'just long enough' and apply a bit of adhesive to hold them in place on the board so they follow the original trace pretty closely, don't you?

>>1433085
Yeah

>>1432870
>Why is it that

I'm not entirely sure of this, but I do have a suspicion:

It's generally much better to have a wire mistaken for hot/live than it is for the live wire to be mistaken for something else. In old installations, the amount of dirt and detritus on the wires will tend to make a wire look black. If it isn't clear immediately, you're more likely to be cautious around a dark wire than a light one, thanks to the fact that whatever "black" wire you're seeing might be hot.

Don't get me wrong, it seems simple enough to simply try and scrape whatever gunk and dist on the wire off to get a better look, but I've had exactly that happen at work. Old, outdoor box that feeds a wall sconce lamp, both wires are a dark grey and were indistinguishable at first glance. Found several boxes like this later, too. It's entirely likely there wasn't any particularly good reason, but this is the best I can think of.

>>1433085
Enameled wire trace repair looks professional
Long ass bodge wires hanging off the board look sexy (when done right)

Both are acceptable.

>>1433097
forgot pic

>>1433098

>>1433052
No, output and input impedances are entirely seperate from impedance matching, at least as far as I understand it. This is just a matter of making the seperate segments of your analog circuit fit together without interfering with the operations of one another.

>>1433052
yes indeed
but dividing an impedance by 100k or more isn't usually super useful

>>1433083
sure, you might have trouble with chattering, depending on your relay. also a lot of general purpose op amps don't work well near the + rail so you'll probably need to invert it all to the - rail instead
are you trying to measure the load current or the supply voltage?

>>1433112
Voltage. Since it will be variable from the source,i want to make sure it cuts off if it jumps too high

>>1433115
ah, Pic related
the diode is a cheap and semi-accurate 0.5V voltage regulator. good enough for what we're doing here anyway. if you want better accuracy you can replace it with a precision voltage reference e.g. TL431 or a Zener diode
anyway we divide the input voltage down so it's well within the range of the adjustment pot, compare, and if exceeded, turn on the MOSFET and pull in the relay

>>1433131

sorry for retardo question, but what generated that cool image

>>1433137
falstad.com/circuitjs

>>1433137
>>1433139
fug
falstad.com/circuit/circuitjs.html

>>1433139
>>1433140

muchas gracias mis amigos

What would be the cheapest way to supply 50 mA at 12 v from outlet power?

>>1433144
use a 12v wall adapter you probably have laying around

>>1433144
Borrow it from your desktop power supply.

>>1433146

( use yellow and black gives 12V)

> the cheapest way to supply 50 mA at 12 v from outlet power?

Stealing first a car battery?
Borrow something from someone you know?
"cheap" depends on your options, and we don't know what you have in your enviroment.

>>1433146

>Borrow it from your desktop power supply.

simple in theory.

hard as shit to actually do. if I have a molex connector it's trivial, so long as my psu has boomer connectors.

>>1433153
Depends on how much junk you have laying around.

.

>>1433144
12V wall wart
really

>>1433153
>not having "molex" connectors laying around
AMP prefers the term "Commercial Mate-N-Lok"

>>1433082
...a whole paragraph of this shit just to say "I don't know what I'm talking about, but I hate Americans."

I think I hear a muslim raping your mother.

>>1433164
The standards are like that because wiring codes got defined by different groups of people.

Granted if you have a panel with both control voltages and mains, nothing stops you from making up your own, printing it out and laminating it into the sidedoor or something.

>>1433167
Some common fucking sense should be used though.

>>1433164
>I think I hear a muslim raping your mother.
I'm not sure my mother had ever heard of a muslim before she died in 1972.

>>1432996
you can with a 10kw fiber laser. We can cut copper up to 3/8" thick.

on a CO2 laser, you are correct.

>>1433168
Fucking this. Just about everything in the electrical industry runs on standards except for the terms governing positive, negative, and earth ground. Color coding of these three basic lines should be standard.

every
industry
gets
this
basic
fucking
concept

Finally finished designing the main board for my 65c02 computer. Hopefully I didn't fuck anything up too badly.

This board has the CPU, 32kB RAM, clock, reset circuit, two expansion slots, and a wait-state generator (since the buffers on the second expansion slot would eat up nearly the entire access time at 14 MHz). I/O devices will go on expansion cards once I decide what I want there.

>>1433144
At least where I live LED light bulbs are cheaper than wall warts, and they have supplies inside.

Plus you get some LEDs

I'm trying to build a computer keyboard with hardware-level macros. I don't care if I have to program it to change it, I'd like to have a button that can press a few buttons with a delay for me. I figure something small like an Arduino would work fine for this. Does anyone have any suggestions on how to go about this?

>>1433353
use an arduino leonardo, literally made for that

>>1433293
How many layers? If it's two layers consider ground pours. Think about where your current is going, you may want to add more decoupling caps and widen power traces.

>>1433355
What do you use to code on arduino? On RPI i can pick any language i want since it runs full on linux, and its costs the same as arduino

>>1433369
Usually C++ with the Arduino libraries, other languages and platforms are available.

You can get the lower end Arduino board clones for about $1.50 each shipped, cheapest RPi is nearly 10 times that if you're in murica, or more elsewhere.

>>1433293
>I/O devices will go on expansion cards
You sure you don't want even blinkenlights on the main board? Or UART?
How are you going to upload the software?

>>1433384
I can buy raspberry zero here for $7.
I have been actually considering buying an arduino tho.
Normally i would use raspberry for 100% of all my projects because having full fledge linux on the board is just soooo fucking comfy. But there is one HUGE downside to RPI. No fucking analog pins. The entire GPIO is digital. What the fuck were they thinking.
If you want to do something as simple as reading a thermistor you need to buy shitty large and expensive analog to digital converter hat or build a gimmicky very imprecise timer circuit with a capacitator.
All i ask is fucking ONE analog pin that can read more than 1 level of voltage.

So i wanted to try arduino for this, but C++ is a huge turn off for me. I hat that bloaty overly complicated crap. I was spoiled by the comfiness and easiness of C# and Python and i just can't go back to that low level crap.

So is there any way to program the $1.5 arduino with python or c#? I don't even need 100% of the functionality, i would use it for simple shit like "if this thermistor resistance is > 1000 ohms then supply 5V to this pin" and such

>>1433386
if you can C#, you can C++ on an arduino, you wont need to write your own classes, you download libraries

>thermistor
use a DS18S20

>>1433389
Can you give an example of a simple arduino c++ code that does what i described above, so i have some picture as to what to expect if i do try to go that route?

>>1433390
Mate it's literally time delays, the kind of shit that you can find in two lines from an arduino tutorial. But some microcontrollers support micropython, I think the ESPs do.

>>1433390
the arduino analog pins measure voltage, so you need a Wheatstone bridge for a thermistor

code:

#define SENS A0 //analog pin
#define THRS 400

void Setup()
{
pinMode(SENS, INPUT);
}

void Loop()
{
if (analogRead(SENS) > THRS)
{
kill(yourself); //any function
}
}

or you can use a comparator/op amp to convert from analog to digital if you just want to threshold the value

>>1433389
>DS18S20
What kind of unicorn is that?
I went to my local electronics store to buy this and the clerk was like "We don't even have that like at all"
So I was like "Do you like have at least any small digital thermistors at all?"
and he was like "sure" and gave me some and i was like "But wait why do they only have two legs" and he was like "..." and i was like "aren't these analog?" and he was like "yeah" and so i left with empty hands wasting a trip and $2 on parking
thanks a lot

>>1433399
Damn that seems really easy, i think i will give a shot in the native language after all

>>1433357
Yeah, two layers. I've already got ground pours on both layers (not visible in the pic).

>Think about where your current is going
You mean the signal return paths? I tried to keep the green layer as clear as I could for those, with shortcuts through the red layer to let them hop over the bigger slots (like below the RAM).

>add more decoupling caps
Where, and for what purpose?

>widen power traces
They're already 10 mils, do I need to go wider than that? The stuff on this board should draw no more than about 200 mA total.

>>1433385
>How are you going to upload the software?
For now, I'm just going to stick a breakout board into one of the expansion slots so I can prototype additional hardware (ROM, UART, etc) on a breadboard. Once I figure out what I'm doing with those, I'll make a "real" expansion card for more permanent use.

Is there a good book for commonly used electronics circuits with examples?

Would be nice if it had some actual illustrations.

>>1433389
I looked at the arduions and there is a bunch, and it seems the one most similar to a raspberry pi is arduino uno so i ordered a bunch to try them out

>>1433389
Can i use arduino uno to program arduino mini, or do i have to buy that ft232 gizmo?

>>1433293
1. waste of a via
2. wtf, just use power fill
3. maybe you should run your busses bottom-side. for 15MHz I don't think you need that strictly continuous of a ground plane especially if CMOS. at 150MHz you would probably want to consider a four-layer board
4. did I mention avoiding unnecessary vias?
5. definitely widen power traces like otheranon said. wider is always better in fact, use power fill on one side and ground fill on the other and get a free tens-to-hundreds-of-pF distributed capacitor
get out of the mindset of using layers for jumpers and loosen up on "bundling" related signals together. that's what the netlist is for. also start thinking in 2.5D
try preferring one layer for significant horizontal lines and another layer for significant vertical lines. you can use the PCIE connector as lots of free vias
summary: 3/10 your board will *probably* work but it could be more compact and cleaner

>>1433353
use a bluepill instead
arduino + HID is a shit

>>1433472
yeah, the internet

>>1433477
uno probably won't do HID

>>1433496
it's async serial. you should be able to figure it out

>>1433411
YM digital temperature sensors? nomenclature is important, mr. capacitatatator

>>1433497
>HID
What is that?

>>1433499
Wait is that like when you can connect usb stuff to the raspberry? I don't really need to connect any usb stuff to the arduini, i have rpi zero for such projects.

>>1433499
USB Human Interface Device. lots of reading to do here
>>1433500
you wanted to make a "computer keyboard", no? if not USB, are you going to use the PS/2 interface to connect to the host?

>>1433501
That was someone else, I manage my raspberries exclusively over ssh, and the arduino projects i want to do will be all pin based.
HID devices are for the weak and complacent.

Guys I'm a retard but I really need help with this. If I needed to power something that uses 1 kw, with a 500 watt generator and a bunch of capacitors, but I only need to power it for a moment, how would I actually store the energy? Do you need to convert it to 1 kw or something? My brain hurts thinking about this stuff.

Are all these grounds connected or separate?
The diagram makes it looks like all those power sources are isolated from each other

>>1433499
>>1433500
HID is an interface standard that USB devices use to communicate with computers and (sometimes) each other (e.g. USB hubs).

If you go the arduino route to make your macro key thing, I'd highly recommend checking out a Teensy board. They're arduino compatibles with 8 bit and 32 bit ARM options. Best part is they have a lot more IO than the 328P in an arduino, more processing power, and are still programmed in C++ with the arduino IDE. And teensy has a built in HID library
https://www.pjrc.com/teensy/

>>1433570
Nah, should work with all the grounds connected and all the +v connections on the same source as well since they're all the same voltage

>>1433569
Charge the caps from from the generator and them let them rip.
Simply add a switch to separate them from the thing you want to power, wait for them to charge and then connect them.

>>1433572
Would it work if the sources were separated?

>>1433574
That's it? No safety equipment needed to prevent explosions or anything?

>>1433575
Yeah, as long as the grounds are connected. Think about the path the electricity must take through the circuit and back to the ground (or Vcc if electron flow is your thing) of the voltage supply—as long as there's a path it should flow.

>>1433576
Depends on the voltage. If your generator is only pumping out say 10V then you can man handle those caps without any fear.
If it pumps out something like 300V, you will fucking die if you touch it.
Some general rules are, never use more voltage for the caps then what they are rated for and never switch the poles (some caps have + and - the longer leg is +)
And if the cap is charged to more than about 50V don't ever touch it.

>>1433576
>>1433579
Also connect them in parallel, this will add their capacitance together.

>>1433504
>FTDIchip virgin
>not a native-USB chad

>>1433570
all connected
if grounds are to be separated from one another, it's customary to use different symbols for each branch

>>1433576
as long as you're using dc. if you're using ac, you're screwed

>>1433579
>>1433580
And use thick enough wires and don't short circuit them. Even at something like 13V where you can safely touch them, if you short circuit them with something conductive you can get very severe burns instantly.

>>1433412
>that seems really easy
there is a reason arduino has become the brainlet's choice for electronics projects

>>1433412
>>1433620
Yeah, they use arduino in middle school STEM classes these days. And if the learning curve is still somehow too steep, they make arduino compatible block code programs where you make a flow chart of what you want it to do and then it generates code for you.
Flashing LEDs has never been easier

>>1433648
Ladder is idiot proof.

You can get chink clones for well under 100USD.

>>1433649
what is that for

>>1433649
Ick, I hate ladder logic and the old shits that still use it to program their PLCs. You can do so much more and with much greater scalability if you're just willing to learn a programming language.

I'll write lines of code all day in any language given before you'll make me use that shit.

>>1433652
A ladder diagram. Ladder logic evolved from relay logic (as in using physical relays as logic devices to control machinery) and its largely used by the manufacturing industry to control assembly lines.
https://en.wikipedia.org/wiki/Ladder_logic

>>1433653
is that supposed be some weird code?
actual code written with letters would be much more simple and understandable
why teach kids this useless garbage when autism friendly languages like python exist?

>>1433660
Oh lordy no, they don't teach kids ladder logic. That's something you'll learn in college when you take PLC classes if you're going for a technical / engineering / manufacturing degree.

I was saying that arduino is piss easy, specifically if you use block code / visual programming, as in pic related

>>1433663
So this is what the machines made the Matrix in.

>>1433660
dude, it's a schematic. you're gonna have to learn symbols anyway. in these cases where semi-skilled labor e.g. maintenance techs needs to read and understand them too, it's better to draw a picture than trust someone to write literate code (see also DRY principle)

What would be simple simple way to implement a battery into a circuit?

Say you have a 9V power source connected to a light bulb. And you also have a 9V battery, how would I add the battery in, so that if the power goes out, the circuit will start drawing juice from the battery?

>>1433663
My point is that they should learn ladder logic to kids, it is cold hard logic with no useless shit on top.

>>1433675
try this. it conducts the higher of the two voltages to the load and prevents the higher voltage supply back-feeding into the lower
caveat: you need the main supply to be a bit higher in voltage than the backup supply, half a volt should be enough
caveat2: the Schottky diodes will drop a little bit of voltage. regular silicon diodes e.g. 1N400x will drop about twice as much for the same current, but you could use them in a pinch

>>1433695
>Schottky
You mean zener? i am confused

>>1433703
the diode bars are S, not Z... admittedly I didn't draw it very well
Schottky diodes act much like regular silicon diodes, but with a lower forward voltage drop and faster action. Zeners start to conduct on a high enough reverse voltage and that is exactly not what you want here

>>1433293

cool board, thanks for sharing. I hope that you'll post the real deal here, it got me curious (and envious!) :)

Damn, I must be more effective with my own time.

>>1433386
> but C++ is a huge turn off for me.

Strong opnion is a sign of passion, use it to develop grit.
You can program it in C, no need to learn C++ quirkness. C# is Satan's spawn, so you'll improve your life by REJECTING C#!!!!

> reading a thermistor
there's some IC who have SPI/i2c.
it depends on your temperature range.

>>1433653
aaah,I want more ladder porn!

It is very fast, very useful at timingsensitive devices. I didn't enjoy the "programming", but I enjoyed the learning phase.

I am glad that I learned that weird beast called PLC. I think that these has aged well, these won't be outdated in 10-15 years.

>want to use use a 12V battery instead of 7.4V on a chinese board with no information or schematic
>decide to check datasheet on the boards voltage regulator to see what the max Vin is
>spend 2+ hours trying to identify SOT-23-6 regulator package
>finally give up and start probing around the board
>hour goes by
>this is leading me nowhere
"fuck it, I'll just give it 13V and see if it dies"
>everything works fine

Well, OK

After a day of trying, I feel like i've wasted my time and inhaled enough lead to kill a bull.

Pic related is the cheap iron I have, I tried soldering back the wire on my headphone that stopped working, problem is, I just can't do it.

When I put solder on the tip and try to press agaisnt the metal it doesn't stick, after 2 seconds the solder dissapears(?????). I did some testing and I think that, somehow, the solder starts flowing to the cooler parts of the iron(the darker part of pic related) so It ends up "sucking" any solder it touches.

Also, I trimmed the wire on my headphone too much, now its really short and I want to replace it, will any copper wire do? It has two wires on each side, one is red and the other is green, don't know if that matters, but im a noob so idk

>>1433903
you're doing it wrong
you need the flux in the connection, not on the iron. you put the iron on the wire and the terminal, then feed solder, then pull the iron

>>1433905
I was doing it that way because I need to hold the wire and iron in place, to hold the solder as well would require a third arm.


Anyway, i'll try that tomorrow and see if it fixes the vanishing solder.

I want to make a BLDC motor controller, that would work with halls, and without them (BEMF detection)
Any recommendation? What MCU should I use? Or should I use one at all?
I'm planning to use it in my e-bike, because stock one is locked and it is shitty due to 20 kHz swithching frequency. Fets are getting hot unless you're not flooring it.

>>1433910
>to hold the solder as well would require a third arm
sure does. they sell those, you know
the equipment challenegd could use a roach clip, binder clip, whatever will hold the wire in the right place for you. iron in dominant hand, solder in the other. you won't need to feed a whole lot. start with a tip clean of solder. try to be quick, get in, get out

>>1433903
>inhaled
>lead
Lead solder does not produce lead vapours. If you solder without any flux then you won't see or smell any vapours; the vapours are solely due to the burning flux.

>>1433903
Oh and I've recently tried to repair a pair of Sennheiser earphones, but the solder just won't wet to the wires no matter how much flux I dump on them. I suspect each strand has an insulation that can't be simply burned off, I might just try cranking the heat up since I've no lighter to use. The strands are too thin to use sandpaper on. You may well have the same issue, but of the 3 or 4 pairs I've tried fixing this is only the first I've had this problem with, so you should try better soldering practices first.

FR1 vs FR4. What is better for prototyping, if I don't have decent drill?

>>1433497
>2. wtf, just use power fill
What does power fill solve there? The point of that section is that I have lines D0-D7 coming out of the CPU and I want them to be in the opposite order when they get to the expansion slot. There are maybe other ways to deal with the issue but I don't see how a power fill would help.

>for 15MHz I don't think you need that strictly continuous of a ground plane especially if CMOS
OK, that's good to know. Most of the other stupid shit I did was due to me trying to keep the ground plane intact as much as possible while staying on two layers only. If this thing really works at 15 MHz then I'll know I can probably tone down the ground-plane paranoia for next time.

And I'll go through and widen the traces a bit before I send it off. Is there a good rule of thumb for signal and power trace widths?

>>1433947
For prototyping I'd still go with FR-4, I find desoldering and desoldering on FR-2 causes traces/pads to delaminate pretty easily. I use a drill-press and 1mm bit, no clamping besides my hands, but I don't think using a handheld drill would be any worse.

Not that I'm your typical PCB etcher, that is. I use nail polish.

>>1433959
But FR1 is way cheaper... At least in mine shithole.

>>1433963
It's all peanuts when ordered from china anyhow.

>>1433970
Hm... Same as FR1 in store nearby.
But I don't like FR4, because glass dust is just horrible. But tracks are stuck really good there.
So FR4 is definitely a way for finished PCB, that is made in china, not in mine empire of dirt.

>>1433952
d'oh, 2 was supposed to apply to 1 and also the oscillator power. you could have routed U3 power etc. from the board edge VCC, on the other side of the output pin, and saved two vias (inductors)
as for 1, it's a bit silly starting a trace on the top and switching sides right away
>opposite order
hence the suggestion to just build a big bundle near the horizontal center of the board and then sort it out at the connector end. otoh I usually just change up my connector pinouts to match easy board routing, where possible. also SRAM doesn't care which order address or data lines are in, you could have flowed thru the IC and put your ground plane perturbations nearer the edge of the board, out of current paths. but be careful between SOIC pins. make sure the board house has soldermask registration/resolution sufficient to cover traces between them to avoid aggravation
>Is there a good rule of thumb for signal
generally, don't use the minimum trace/space unless necessary. the less you push the limits of the manufacturer's process, the better yields and fewer delays you get
KiCAD has a decent PCB calculator. Pic related contains the IPC standard formula for maximum average trace current if you want to math it out manually. try to keep temperature rise not more than 5°C (do not exceed 10°C) on FR-4 for actual average current
when you get into faster or differential signals, impedance starts to become important and you will need closely matched trace widths and constant spacing to avoid perturbing high-speed signals. the PCB calc has your back there too
>power trace widths
generally, avoid running traces to a pad that are as wide as the pad itself (delamination risk increases), or neck them down just before the pad. otherwise, wider is better. board houses would rather ship copper than etch it
now that I understand your design strategy I'll give it a 7/10, if you make a faster board it will probably work nicely

Gimme some cool idea for project. Pls

>>1433986
enjoy

>>1433989
Thanks.

>>1433293
>Finally finished designing the main board for my 65c02 computer.


i did something similar a while back, but i gave myself a huge advantage by using a hardware design that closely followed a commercial 8-bit machine. (who, in turn, essentially copied the reference design of the CPU's manufacturer). so what's the advantage? it means i can pop in a EPROM with the commercial machine's code and have access to a ton of useful subroutines for I/O, graphics, sound, keyboard decoding, math functions, including floating point, and even a full BASIC interpreter if i want it. it would have cost me thousands of hours to replicate all that.

>>1433293
>Finally finished designing the main board for my 65c02 computer.

i did something similar a while back, but i gave myself a huge advantage by using a hardware design that closely followed a commercial 8-bit machine. (who, in turn, essentially copied the reference design of the CPU's manufacturer). so what's the advantage? it means i can pop in a EPROM with the commercial machine's code and have access to a ton of useful subroutines for I/O, graphics, sound, keyboard decoding, math functions, including floating point, and even a full BASIC interpreter if i want it. it would have cost me thousands of hours to replicate all that.

i can also adapt third party software like assemblers, monitors, debuggers, and terminal emulators to work on my machine, once again saving me tons of work.

>>1433972
Just dump some water onto the bit while you're drilling, no dust. Also, in terms of price, I found a pack of 50 4x3 inch double sided boards for $20, with prime shipping too. Don't know where you are, but at my local electronics store copper clad is also expensive as hell, and that's near a big city in the US.

What are the frequency limitations of Class D amps?
I want to drive a transducer at 30kHz, I was looking at boards based on TPA3116D, but I can't find any information if they are capable of handling 30kHz

>>1434038
That IC runs an oscillator frequency from 400kHz to 1.2MHz so you'll not run into any hardware-borne incompatibilities with 30kHz, and on the datasheet the example circuits have low-pass filters set to 50-60kHz, though you'd need to check the model you buy for its filter values. Even if they're set too low you shouldn't have any problem replacing the caps or inductors. But the circuit diagram might have something along the lines of an input filter at the gain control section with those capacitors. You should be fine, since they look like they just have to do with oscillator-frequency feedback, since they're connected to the unfiltered output. Unless anyone else can find a problem I'd go for it.

I guess you can't just run a square wave into your transducer?

>>1434040
I haven't tried, but I imagine the results would be worse. I want to use it for ultrasound rangefinding, ideally with at least 30-40m range (one-way, with a receiver on the other end).

Is this logic level mosfet?
https://www.vishay.com/docs/91054/91054.pdf
I am confused be the sheet, at one point i says Vgs is 5V for logic level
But then in the specs it says VGS = 10V

>>1434048
That is odd. It specifically labels the GS threshold voltage as 2-4V, while the 10V one at the top of the datasheet is just V_GS, so I'm leaning towards yes, but the V_GS_MAX is also specified as 20V, so I'm unsure. I'd take a look at another datasheet by someone else (if possible).

>>1434054
Shit, turns out it's not.
But what I don't get is they sell another not logic level mos, in this pack here
And they say you can trigger it with arduino, which is about 5V.
Does that mean I would have to connect extra power source just to open it?

>>1434062
I have a very similar module, not sure if it's exactly the same, but that one works with 5V, directly from the arduino. I imagine that this is probably the same chinkshit.

>>1434066
That is weird, since I checked the fet in it and it is not a logic level.
Or it's simply full ovechink mode and when they were ripping the western product off, they just printed a different name on the fet, even though it is logic level, or they just half assed the spec sheet, because they just don't give a fuuuuuck.

>>1434048
A MOSFET has no threshold voltage in the sense that at a Vgs of 2..4 volt a magic contact suddenly closes turning it from OFF to ON. The data sheet explicitely specifies what threshold voltage really means in this context and how to measure it: Vgs=Vd and Id=250µA. The operational conditions can be found in the diagrams elsewhere in the data sheet.

Pic: In reality a MOSFET has no threshold at all. Vt (2.1V in this example) only marks the voltage where the exponential characteristic (at low currents, right diagram) changes to the quadratic one for higher currents (left diagram).

>>1434070
TTL Voh is a bit lower than CMOS Voh, so the term "logic level" is a bit vague and dated
better to look at the Vgs-Ids transfer characteristic graph in the datasheet and see if the MOSFET will pass the required current at the required drain-source resistance when given the available gate voltage
better still, choose a MOSFET whose Rds(on) is specified at your available gate voltage

>>1434015
Well, part of my plan for this machine is to add custom hardware for interfacing with as many "modern" peripherals as I can manage: PS/2 keyboard, SD card, maybe USB HID & mass storage if I can figure that out. So I'll be writing a lot of my own drivers no matter what.

>>1434164
but it could still be nice to have a framework already there to design your drivers around and plug them into
>PS/2, SD card
you should definitely lern2fpga for that. an ultra-low-density device like the Lattice iCE40LP384 would be plenty for a UART, SPI, I2C, or any other basic bus-controlled serial interface, maybe even two. you could also cram a DMA controller into the larger chips and still have LUTs and pins left over for glue logic and address space decoding. some of the larger series in the range have hard-core SPI/I2C master and slave controllers built in, for which you need only build a bus interface
it would also be cheaper than doing it discrete logic style, unless you need 5V tolerance on the bus. if you're designing a peripheral-heavy system you'll probably want more than two interrupt lines, maybe even vectored interrupts
>USB
the FTDI Vinculum2 would be a good low-cost chip for that. it's actually a microcontroller in and of itself, but has some precompiled firmwares that serve as a general purpose USB host and an Intel-style parallel slave interface in the form of a FIFO to send that firmware commands and data. some of those firmwares also have built in FATFS support for USB mass storage class devices

How much heat is one watt?
Put that in a practical example for me.
Like how hot will a coin get if it has to dissipate one watt?

>>1434212
that depends on many other factors, such as how well the coin gives off heat to the atmosphere
fortunately, heatsinks are often rated in degrees Celsius per watt of dissipation. if you apply 10W to a device connected to a 5°C/W heatsink, the device at the point closest to the heatsink will be 50°C hotter than ambient
you might want to look up free physics and/or thermodynamics textbooks to get some idea of what you're dealing with

>>1434218
I just want to be able to judge a size of heatsinks. Like if i have a linear voltage regulator how big of a heatsink should i plan for to dissipate values like 1W, 10W, 20W etc

what is the best cheap chinese oscilloscope I could get off aliexpress?