/ohm/ schottky diode, the most aesthetic circuit symbol edition

[Anonymous No.1471126]

wise, aged old thread: >>>>1466111
→

>RULES (draft, comments welcome):
0. Mains wiring goes to /qtddtot/. PC assembly to >>>/g/.
1. Know the law. You are assumed to understand Ohm's Law and Kirchhoff's circuit laws. If not, try /sqt/ or /qtddtot/
2. RTFD. Re-read all documentation/datasheets pertinent to your components/circuits before you ask.
3. Pics > 1000 words. Post schematic/picture/sketch/9001.5 hours in MS Paint with all part numbers/values when asking for help.
4. Read the problem statement before replying.

>I'm new to electronics, where to get started?
It is an art/science of applying principles to problems. Find problem, learn principles, design and verify solution, build, test, post results, repeat.

>Problem ideas:
http://adafruit.com
http://instructables.com/tag/type-id/category-technology/
http://makezine.com/category/electronics/

>Principles (by increasing skill level):
Mims III, Getting Started in Electronics
Platt, Make: Electronics
Geier, How to Diagnose & Fix Everything Electronic
Kybett & Boysen, All New Electronics Self-Teaching Guide
Scherz & Monk, Practical Electronics for Inventors
Horowitz and Hill, The Art of Electronics

>Design/verification tools:
LTSpice
falstad.com/circuit/circuitjs.html
NI Multisim
CircuitLab
iCircuit for Macs
KiCAD (pcb layout software, v5+ recommended)

>Components/equipment:
Mouser, Digi-Key, Arrow, Newark (global)
RS Components (Europe)
eBay/AliExpress sellers, especially good for component assortments/sample kits (caveat emptor)
Your local independent retail electronics distributors
ladyada.net/library/procure/hobbyist.html

>Related YouTube channels:
mjlorton
paceworldwide
eevblog
EcProjects
greatscottlab
AfroTechMods
Photonvids
sdgelectronics
TheSignalPathBlog
BigClive

>Li+/LiPo batteries
Read this exemplary resource first: https://www.robotshop.com/media/files/pdf/hyperion-g5-50c-3s-1100mah-lipo-battery-User-Guide.pdf
>I have junk, what do?
Take it to the recycler.

[Anonymous No.1471129]

should I major in Computer Engineering or Electrical Engineering - Subspecialty VLSI Design?

[Anonymous No.1471130]

>>1471129
I guess I should be asking, which has better prospects?

[Anonymous No.1471133]

>>1471126
geez, no need to wet your pants so quickly, the old thread still has days left

anyway this thread's digits brought to you by the Allegro A1126 chopper-stabilized omnipolar Hall effect switch

[Anonymous No.1471136]

>>1471133
>Brought to you by a Allegro Hall-effect switch
Currently working on a A3144 project I heartily approve.

[Anonymous No.1471138]

>>1471126
>Them rules
Looks more in the spirit of /g/ than /diy/

[Anonymous No.1471140]

>>1471138

and OP seems to have confused "problem" and "project".

Nobody reads the OP crap anyway.

[Anonymous No.1471142]

>>1471138
we hit bump limit quickly enough without questions that somehow land here because omg there's a wire somewhere

[Anonymous No.1471144]

>>1471140
I'd say that we get more problems without projects than projects without problems. When was the last time someone came along to /ohm without any questions just to show off his cool clock? There's almost always a problem to be solved, and that's a good thing.

[Anonymous No.1471160]

>>1471144

are you actually saying that people go to adafruit looking for problem ideas, like OP has listed?

k.

[Anonymous No.1471219]

>>1470973
What suit maker, this is for a prop weapon. I need to light up a legnth of acrylic that's about 2ft long and 3/8 inches thick, I CANNOT use LED strips for this application because there is nowhere for me to hide them. So yes I can use 2 AA batteries with it and skip the heat sink and be fine?

[Anonymous No.1471243]

hope i am in the right place:
my gpu fan started grinding about a week after my warranty expired. tore it apart, found the exact part number, ordered it from shanghai, and now it's here. the plug into the card is different. the old one had 4 wires (black, yellow, green, blue) and the new one has only 2 (black and red) and the male bit is much smaller. anyway to (safely) splice the new wires into the old connector?

[Anonymous No.1471277]

Will a stepper motor be damaged if i grab it with my hand while it's spinning and hold it so it can't spin?

[Anonymous No.1471286]

>>1471277

in general, no. if it is being driven by a current mode driver then it's almost impossible for it to overheat, assuming the current limit is appropriate. If it's being driven by a high voltage and it always ramps to a high speed then stalling it could cause it to overheat, but this is not a typical method of driving steppers.

[Anonymous No.1471292]

>>1471243

look at what pins on the mobo the red/black are connected to on the old connector, and connect the red/black to the same pins on the new. (that leaves two pins on the mobo unconnected.)

also, you can look to extending the life of the old fan. removing the central label often gives access to a bearing that you can drop oil into to silence the noise.

[Anonymous No.1471295]

>>1471286
Ok thats good. Also how come i can't turn the motor easily when i disconnect the power? since the coild inside aren't powered, shouldn't it be easy to turn?

[Anonymous No.1471296]

>>1471295

if the wires are not shorted through diodes or any path then most steppers can be easily turned by hand. there is some resistance because the magnetic field is being reversed in the stator lams, and there will be some hysteresis and eddy current losses. Steppers also usually have what is sometimes called cogging torque or detent torque which is the holding torque at each step position, and this will resist rotation at low speeds but tends to average out at higher speeds.

If you short the coils then there will be more resistance to rotation because a permanent magnet motor acts as an electrical generator.

[Anonymous No.1471298]

>>1471243

you can look to extending the life of the old fan. removing the central label often gives access to a bearing that you can drop oil into to silence the noise.

as for the the new fan, splice the black to the black on the old connector, then try the red on the yellow, or each of the other 2 colors if that's no-go.

[Anonymous No.1471299]

>>1471160
I'm saying that if they don't have a problem they usually don't feel a need to post here.

[Anonymous No.1471304]

>>1471299
on the one hand, I kinda wish they would
on the other hand, I've seen blogs devolve into an endless series of binary wall/desk/etc. clocks

[Anonymous No.1471337]

>>1471127
>haven't seen it, but you could do worse than drilling/tapping PC CPU coolers according to their rated dissipation
You're right of course. But when i make my final diy unit, I would like it to have the sort of heatsink where the fan blows down all the fins and out the case, ideally. seems finding heatsinks is a journey in itself :/

[Anonymous No.1471339]

>>1471296
Normally generators have high resistance when open, and less when shorted. They're easier to turn when shorted. Are you sure you're not backwards?

[Anonymous No.1471340]

>>1471339
>They're easier to turn when shorted.
They're harder to turn when shorted because they're trying to do maximum work.

[Anonymous No.1471345]

>>1471339
Not him but why would they have less resistance when shorted? Lenz's law tells us that there'd be an opposing magnetic field thus making it harder to turn when shorted

[Anonymous No.1471348]

>>1471337
you could go full memelord and use water-cooling, and put the air flow wherever you like

[Anonymous No.1471350]

>>1471339
>Are you sure you're not backwards?

I have designed them and tested them and I know how they behave. feel free to say that a generator under heavy load (low ohms) is easier to turn than a generator under no load, open windings.

where do you morons come from???

[Anonymous No.1471358]

>>1471350
what if it's all the same shitposter?

[Anonymous No.1471366]

>>1471358

there's a naive part of me that stupidly thinks the /ohm/ general self-cleanses itself of retards.

but then it's 4chan. possibly one of the best parts of 4chan, but still.

[Anonymous No.1471370]

>>1471348
>memelord and use watercooling
Bro i'm trying to make a diy electronic load lmao
If i used water cooling it'd be because i was proficient at making whatever it is AND there was a huge heat load. Then definitely i'd go full memelord kek

[Anonymous No.1471395]

>>1471370
For my variable load I plan on just epoxying the load TO-220 transistor halfway into a piece of vinyl tubing and hooking up a tap to one side. Might have it on a small heat-sink which then is shoved into the tubing but I don't think it would make enough of a difference.

>>1471366
Any other good parts of 4chan you'd recommend? I find /g/csg pretty comfy.

[Anonymous No.1471398]

>>1471395
Just running straight water past it? I guess that's simple enough
I've tried doing the whole, "long length of wire immersed in water" as a dummy load. That was kinda cool. felt the water heat up slowly. Just think its a better learning experience to make this stuff, and to make it with certain options in mind. constant current. constant resistance. maybe even constant power? I understand there's a way to mulitply values with op amps but i haven't looked into that
Don't know how to do constant resistance either. Sensing the load voltage, dividing it by the sensed load current, then having feedback to change load current so that the divided value is the same as some set resistance value? That's how i'd do it naively, but not sure if there's a more elegant way to do it
The only thing I'm using arduinos for is to do read-outs and put them on a small oled screen. On both it and my power supply project
And then electronic loads get more complicated if you want to make it be able to do variable loads and such. Course at that point you might as well buy an electronic load
>tfw have to learn to make a waveform generator to make an even more impressive electronic load

[Anonymous No.1471406]

>>1471398
I planned on just using a FET itself as a linear regulator, with feedback directly from a current shunt + op-amp. I guess with a wire immersed in water you'd PWM it? I can't imagine the spikes of 20A+ would be very nice when it comes to stray inductances. With using a FET as a linear regulator, a constant-voltage variable load would explode really easily and a constant-power or constant-resistance variable load would likely require some digital work, but current-limited is pretty easy this way. Having a system that is constant voltage and switches over to constant current when it gets high enough is probably possible with analogue components.

This linear method is also for DC only, not sure how you'd do current feedback on AC.

[Anonymous No.1471407]

>>1471406
Wait what happens when I feed a current mirror into one of these current sinks with different current settings?

[Anonymous No.1471411]

>>1471406
I meant just testing poer supply regulation with just a long thin wire that i dumped into water. I definitely don't have like 1 ohm resistors or sub 1 ohm, so i just used like a foot of coiled up magnet wire as resistance, immersed in water
What am i PWMing here? I think i'm lost sorry.

My little anecdote was just me trying to say that i tried wire in water as a dummy load because i don't have an electronic load, and that I want to make an electronic load so that I don't have to fiddle with wires in water (and whose parameters can't be set with a knob or something)

[Anonymous No.1471415]

>>1471411
Also, having a lot of op amps series'd with each other.. wouldn't that introduce a lot of phase shifts which would cause the whole setup to start oscillating instead of giving negative feedback?
Control theory is another thing that is esoteric that I sorta need to understand. Some situations when i had my power supply mk2 where the output goes into oscillation and I can't understand how/why

[Anonymous No.1471430]

>>1471298
thanks fren
i guess i could try the lube idea, but i think a powder type would be better/ less chance of drips causing damage. any recommendations?
i dont know if it would be a problem to just try to find the correct jumpers and leave the other 2 bare. also, what could the other 2 lines be?

[Anonymous No.1471431]

Noob here. Trying to find babby's first oscilloscope. I want something that will last me through my self-learning phase and will still be very useful once I start doing more advance stuff. I have a budget of <$400. After doing some research I've found a couple units that are recommended. What exactly should I be looking for though? One has "200mhz bandwidth 2 channels" and the other has 50 with 4 channels. Which one would be better? I know nothing about these things. I mostly want one so I can learn how to use one, not because I currently have a need for one.

[Anonymous No.1471435]

>>1471411
Since the load resistor is a constant resistance, you increase it's effective resistance by turning current to it off and on with a MOSFET. 50% duty cycle = 50% power = 50% resistance. But it requires filtration capacitors to smooth out the current from the perspective of the power source, plus I'd use a current shunt before that capacitor to provide current feedback instead of just controlling it by duty-cycle since the resistance will drift.

[Anonymous No.1471447]

>>1471431
Get the lowest cost new digital oscilloscope. It'll probably be good enough for all your needs past self learning phase (i bought a second hand analog scope)
You'll only ever need the expensive big boy scopes if you're doing something really specific or you're a legit engineer (i'm assuming you're remaining diy)
Or maybe doing radio frequency shit
Beyond that you are probalby good with any modern DSO new even the lowest models
Someone correct me if i'm wrong

>>1471435
But I don't want pulsed load. I want constant power, not effective power

[Anonymous No.1471454]

>>1471435
>>1471447
sorry i kinda glossed over the last part of your post. nvm some of the shit i said in response >>1471447

[Anonymous No.1471487]

>>1471454
The PWM method is likely an efficient way to do things, but the feedback may be a little shaky since the voltage across the feedback shunt will have droops and spikes. Putting a low-pass filter on the feedback could help, but that means at least 2 op-amps I believe.

The issue with using the MOSFET itself as the power wasting element is that a single FET package has limited power capacity and needs a heat-sink. I for whatever reason was stuck thinking that increasing the shunt resistance and having it radiate a significant portion of the heat was a method you could take to reduce heat-sink requirement, but not a good one since the heat would cause the resistance to change. But I just now realised that you could have the current shunt at the source and a power resistor at the sink and take up up to half the power that the FET would otherwise have to dissipate. These things can also easily be stacked in parallel, since they're current-balanced.

One potentiometer, one resistor (to bias the potentiometer), one op-amp, one MOSFET, one current sense resistor, and maybe a power resistor for a single unit. Plus heat handling. But the op-amp needs to be able to go right down to the -ve rail.

[Anonymous No.1471510]

>>1471487
Don't big boy electronic loads just use mosfets though? Naively I can see how using pwm like you said could be made to produce arbitrary loads. But if so, why wouldn't that be more common in professional electronic loads? It sounds like it'd save money and be more compact

[Anonymous No.1471529]

>>1471510
>Don't big boy electronic loads just use mosfets
Yes I think they do, but splitting up the thermal load will help me to avoid the need for active cooling. I only plan on burning maybe 15-20W from a USB port with the one I'm designing.

I think the PWM version would just have too much noise when compared to a linear model and it wouldn't perform any better. If you have a tank of water at the ready or a matrix of wire such that it will sink heat to the air efficiently then there's nothing particularly wrong (aside from possible inductive spikes at high-current), but just slapping a fan on a heat sink is probably a more compact method regardless of what's attached to it. The only variable loads I've looked at are the aliexpress ones for USB, they have a ~5cm x 5cm square heat sink and a fan of the same size, and they look to handle 10W with ease. Actually a 20W LED would need less heat-sinking than a resistor since a significant portion of its power will be radiated as light, so if you ignore the whole blinding thing it's actually a good way of sinking heat. Perhaps an IR COB? Would probably be much worse for your eyes though.

[Anonymous No.1471531]

>>1471529
I looked up this idea and its already been discussed on EEVblog
Anyway, that thread aside, the pwm idea would only allow you to create loads that are the same as the combined wattage of whatever power resistors you used, or lower. Which is basically what you'd do with operating mosfets in their linear region anyway
But with the caveat that it's more practical because you don't need extra filtering and monitoring in order to smooth out current / voltage from the PWM
Someone in the thread mentioned that mosfets are cheap compared to big beefy power resistors so I think the mosfets in linear region method is more cost effective too
I'm sure engineers have thought of this stuff before already and decided it was not as effective nor cost effective. It sounds like an interesting project though

[Anonymous No.1471532]

>>1471531
I can agree on that.

[Anonymous No.1471543]

>>1471531
> Someone in the thread mentioned that mosfets are cheap compared to big beefy power resistors
Power resistors are expensive because they have very low temperature coefficient (little change in resistance over a wide temperature range). MOSFETs don't have that property, and if you don't need it a piece of nichrome wire will be cheaper than a MOSFET.

Also, bear in mind that nominal MOSFET power ratings tend to be bordering on fiction. Datasheets typically give the power rating for the bare die, but given the junction-to-package thermal resistance you'd have to immerse the package (and heatsink) in LN2 to actually dissipate that much power without exceeding the maximum temperature.

[Anonymous No.1471549]

>>1471543
It's okay I look at SOA graph a lot
Also this is why feedback is important

[Anonymous No.1471560]

>>1471543
Temperature coefficient doesn't matter since you're not relying on it having constant resistance at all. By putting linear feedback on a low-power shunt resistor (which does need to have constant resistance but doesn't heat up much at all) your MOSFET will automatically decrease its effective resistance as the resistance of the power resistor decreases.

That's assuming a circuit like >>1471487. So you could just buy a power resistor with 10% tolerance (which surely exist) and be fine with it.

When it comes to maximum power ratings, I think using the thermal resistance rating from junction to ambient along with the junction's maximum temperature is sufficient. I'm getting 62°C/W and 175°C (might be a bit optimistic) from the IRF9Z24N's datasheet. Assuming 30C° ambient, that's 2.3W (no heat sink). With a thermal resistance of 10°C/W (should be obtainable with a reasonable heat sink) it's about 15W. It's impossible to get this over 45W with a perfect heat sink because of the internal thermal resistance.

It's about $1.50 for a 1Ω power resistor on alibay, but powering it linearly with a FET would mean to get twice the resistance you'd have a significant amount of power dissipated across the FET (1/4 the maximum power of the circuit) so in the case that you can't get enough heat out of a single FET it may be more economical to go for the PWM+filtering route. But since it's 1/4 power, you could use a tiny heat sink on the FET and no fan and a 2.5Ω 10W power resistor, and only have to dissipate 2.5W out of the FET as a maximum. But I'd boost the maximum power for the FET to over 3W to account for thermal drift of the resistor. Yes it does get to 30°C sometimes, deal with it.

[Anonymous No.1471568]

>>1471560
Not guy you're responding to but I've never really took the time to figure out that thermal resistance stuff
For the IRF9Z24N, do I just add the junction to case then case to ambient? Giving me 3.8C/W
So now if I have 1W passing through the mosfet while it's heatsinked, I should expect it to be 3.8 degrees above ambient? I know this is a simplified mode but is that roughly the idea?

[Anonymous No.1471574]

>>1471568
It already has a junction to ambient there, along with junction to case, but this datasheet isn't stating it in terms of junction to case and case to ambient separately. You'd just get case-to-ambient by subtracting the 3.3 from the 62. It has what you're more likely to find useful, you're never going to need the case-to-ambient rating without the junction to case being incorporated in there. So as far as I know you take the total thermal resistance, say the 3.3 J>C, 0.5 C>HS, and maybe 10 HS>A and add them (HS is heat sink, others should be obvious) up to 13.8C/W. This means that from the ambient temperature, every watt of heat the device is dissipating will cause the junction to heat up by 13.8°C. If you're wasting 3W then the junction will be 41.4°C above ambient, say 66°C. If your maximum junction temperature is 80°C then you subtract the ambient, say 25°C from that, giving a ∆T = 65°C, which you divide by the thermal resistance to get maximum power = 4.7W. Hope this helped. Also hope it's correct.

>tfw looked at an SMD resistor
>it said TEE on it
>what the fuck does that mean
>start testing
>turn it over
>it's just a 331 = 330Ω
oof

[Anonymous No.1471579]

>>1471574
>Also hope this is correct
If not it seems close based on some shit I've skimmed
I'm lost though. Basically I'll hit TJmax if there's single digit wattage passing through the mosfet
But then how is it that they don't explode when there's double digit wattage being passed by the mosfet in electronic loads? This is all very confusing

[Anonymous No.1471590]

>>1471579
By using heat sinks. Instead of the massive ~60°C/W that you get from case to ambient you get something closer to 10°C/W from case to heat sink to ambient. If you look up the datasheet for a heat sink it will give you thermal resistance values, probably for both in unmoving air and with a fan.

[Anonymous No.1471592]

>>1471590
Okay I see now
But that is another problem - where do i find some heatsinks?
I'm thinking something like pic related
Two of them, with fins pointing at each other, so it's like a tunnel that's filled with fins with the flat faces facing outwards.
Mosfets would be bolted onto the flat faces, and a fan would blow down the "tunnel" and out the case
There's heatsinks like that on banggood, but where can I find some that are similar to that, but which are machined so that they can be bolted onto a PCB and have holes for mosfets already drilled into them etc?

[Anonymous No.1471593]

>>1471592
There are special tunnel-shaped heat sinks just for that. The circular holes on the sides are for screwing into, but not very easy to put into a PCB. The aluminium ones will all be impossible to solder onto a board anyhow, but I imagine there are boltable ones or ones with tinned leads to solder onto.

[Anonymous No.1471596]

>>1471593
These are good for PCBs but are too small for using as a tunnel. Still, since the contact area means most of the heat will be in the centre of the heat sink; it won't spread to the sides too quickly, it might be better to go with a small heat sink like this and have a fan nearby, at least for a single FET. Planning on using two or more FETs?

[Anonymous No.1471597]

Oh god what the fuck. This looked so much bigger in the photo, i thought it was same size as an arduino. I can barely even see this cunt i will NEVER be able to connect any wires to this FUCK

[Anonymous No.1471598]

>>1471593
>>1471596
>Planning on using two or more FETs
Definitely
For electronic load I'll need a few to distribute the current across a few FETs and therefore the heat load across a bigger area. I mean at least need 2. Maybe even 4 or more
And since it'll be an electronic load, I'll definitely want tunnel assembly. It doesn't have to be inordinately long. But definitely long enough to comfortably fit and cool 4 - 8 FETs, TO-220 or TO-247
And yeah I'm just going to bolt them to PCB. But beyond that I lack the tools to properly drill holes and such with a heatsink like in my post >>1471592

>>1471596
your pic related is probably enough for a low power linear regulator supplying power to some op amps or such for sure though. Just I'm hoping I can sink like 100W, and ideally some day up to 500W in a more improved design or something

[Anonymous No.1471600]

>>1471597
You'll be fine. Just do it very carefully. Magnifier helps, maybe "helping hands" (the thing with a magnifying glass and some alligator clips to hold shit)
Probably the time to use some of the more narrow soldering tips

[Anonymous No.1471604]

>>1471600
No way i can solder that without shorting half of the contacts together. The only way will be a very thin solid wire tied into a loop through those holes

[Anonymous No.1471605]

>>1471604
Where's the can-do attitude man?
You'll be fine. If you have flux in a syringe or in a tub, slather some on all the contacts so that the solder won't bridge as easily

[Anonymous No.1471606]

>>1471597

Nigga are you serious? That's some entry-level shit. Pic related I can solder that hand, and I'm not even particularly good. You should have no trouble soldering to even relatively fine-pitch through-holes (which those aren't) with even minimal practice.

[Anonymous No.1471607]

>>1471597

Nigga are you serious? That's some entry-level shit. Pic related I can solder by hand, and I'm not even particularly good. You should have no trouble soldering to even relatively fine-pitch through-holes (which those aren't) with even minimal practice.

Just go HAM on it with the flux and you'll be fine.

[Anonymous No.1471609]

>>1471607
Not him but how do you solder that without a heat gun?
I would imagine being so light that the surface tension of solder would make it cling to the solder of the soldering tip

[Anonymous No.1471618]

>>1471609
Yeah, sure, they like to cling to soldering iron.
You can use forceps or a needle to keep it in place while soldering.

[Anonymous No.1471620]

>>1471607
Yeah sure i bet you get ALL the pussy with you're uber solder skills, but i am not quite so gifted so i struggle

[Anonymous No.1471622]

>>1471598
If you're going for 100W then since a single TO-220 package shouldn't be dissipating more than ~40W, so using 4 of them spread around a single cooling array would be a way to go, but for 500W I'd recommend looking for a more elegant solution. Very high-power package sort of thing, sort by maximum power for MOSFETs on mouser/digikey and see what packages come up. The transistor itself doesn't matter much, just that it has a threshold voltage in the range of the logic circuit.

[Anonymous No.1471626]

>>1471609
>the surface tension of solder would make it cling to the solder of the soldering tip

It wants to. You have to pin it down with whatever, usually your tweezers.

Real talk: While I _can_ do it with an iron, I won't unless I have a really good reason to do so, like heat-sensitive parts nearby or something.. Paste and a heat gun is just way easier at these scales.

>>1471620
>you get ALL the pussy with you're uber solder skills,

They aren't "uber solder skills". That's my point. It's simply not as difficult as you're making it out to be in your head. Just try it. I seriously doubt you'll have any trouble if you've done any soldering in your life before.

Video related. That 0402 on my finger is the _second largest package_ pictured on this guy's. An 008004 is so small it's an INHALATION HAZARD, FFS. If this guy can manage soldering something small enough to be mistaken for a literal mote of dust, I'm sure you can manage a through-hole pad and some wire if you give it an honest effort.

https://www.youtube.com/watch?v=P72U2yHWKSk

[Anonymous No.1471628]

>>1471618
>but for 500W I'd recommend looking for a more elegant solution. Very high-power package sort of thing, sort by maximum power for MOSFETs on mouser/digikey and see what packages come up
When I (probably) make a unit that dissipates that much power, I'd probably use some TO-247 packages
And probably like 12 - 24 of them
Is that what you mean by more elegant?
I imagine that's largely the more practical solution. I understand those mosfets are quite pricey though. For example the IRFP250N

[Anonymous No.1471629]

>>1471626
>Real talk: While I _can_ do it with an iron, I won't unless I have a really good reason to do so, like heat-sensitive parts nearby or something.. Paste and a heat gun is just way easier at these scales.
Agreed
Also dem nice solder connections. Helps to have a nice setup for working with tiny SMD components though

[Anonymous No.1471632]

If a motor has a gear on it that is 50% smaller than the gear it's connected to, does that mean the larger gear outputs twice as much force?

[Anonymous No.1471636]

>>1471431
You can easily unlock 100 MHz in the scope in your picture, that's why everyone buys them. Just google the name of the scope and "licence unlock" or something like that. Unless you need to work with very high speed signals (like above maybe 20 MHz for analog and 40-50 for digital) I'll just get the rigol. Having 4 channels is nice.
Also if you want to work with digital get at least one of those cheap 8-10 dollars chinese USB logic analyzers.

[Anonymous No.1471637]

>>1471609
You can do it with a $9 walmart soldering iron
>use flux
>tin the tip of the iron with the tiniest amount of solder
>hold it in place using tweezers with one hand
>touch it with the tip
>should be held in place enough to do the other side and then add more solder afterwards if necessary

just practice on some old pcb. its way easier than you think especially with flux
for the wifi thing you really can't mess up if you use flux and dont go crazy with the solder

[Anonymous No.1471639]

>>1471637
>for the wifi thing
I'm not him, i said
did you even read the post?

[Anonymous No.1471652]

>tfw you can't even bodge wire a 14nm chip with a heated rebar and your moms janky hairdryer

[Anonymous No.1471655]

>>1471632
At half the speed, yes. Though the correct terminology when dealing with rotational systems is angular velocity or more commonly angular frequency, and torque. Also this isn't a problem for /ohm/.

[Anonymous No.1471669]

>>1471126
I'm a newbe :^)

I wanted to make a 3 position rocker switch (I O II) splitter for computer 4 pin molex. I wanted I to turn 1 device on and II to turn another one. How should I do that? Molex has 2 grounds and 5v and 12v pins. It's for HDDs so I'd like to be safe.

[Anonymous No.1471673]

>>1471669

switch 5v and 12v, hardwire the grounds

[Anonymous No.1471678]

>>1471673
I'd just use a 2-pole switch instead

[Anonymous No.1471680]

>>1471678

what's the difference?

[Anonymous No.1471688]

>>1471680
you don't need 2 switches...

[Anonymous No.1471697]

>>1471688

how does the dpdt rocker need 2 switches; it seems functionally to be the exact same thing as the toggle switch; six terminals.

[Anonymous No.1471700]

>>1471697

you can only see 3 in the photo. like when you see a sheep sideways. you only see 2 feet. how can you be sure it has the one other?

[Anonymous No.1471704]

>>1471700

???

[Anonymous No.1471705]

>>1471704

i can confirm 4 pins visible, one more than previously.

[Anonymous No.1471707]

>>1471705

I give up. You win. OP needs to avoid DPDT switches at all costs.

[Anonymous No.1471719]

What's the difference between Multisim Student Edition and Multisim Full/Regular?
You can't beat $42 compared to $2k+, so what's the catch

[Anonymous No.1471787]

>>1471430

oil wont hurt a computer. you can even dunk the motherboard in oil to cool it.

the 2 extra pins are used to report the speed of the fan, and to control the speed.

[Anonymous No.1471794]

>>1471719
Can't be arsed to see what NI says, but
>student edition
means that you have to be a student to buy it and use it. NI obviously hopes that you'll switch to the full version when you aren't eligible to the student version anymore.

[Anonymous No.1471801]

>>1471794

very helpful answer. i'm sure some people were unsure what the word ''student'' meant in ''student edition'', and now we know.

[Anonymous No.1471807]

>>1471801
Why did you ask if you already knew?

[Anonymous No.1471864]

>>1471807

god, you're thick. it's obvious the dude wanted to know if there were restrictions, such as on number of components, on libraries, on macros, on file exports, on having watermarks when you print, and such like. he was not looking for dictionary definitions.

[Anonymous No.1471944]

I was asking about soldering advice a couple days ago.
ended up buying a TS100
It's not pretty but I fucking did it
thanks lads

[Anonymous No.1471947]

>>1471944
based. i solder shit all the time and if i tried that it'd be bridged in 5 places.

[Anonymous No.1471949]

>>1471944
>TS100
holy shit why
The TS80 is objectively better

[Anonymous No.1471951]

>>1471947
>if i tried that it'd be bridged in 5 places.
Use solder braid to remove excess.

[Anonymous No.1471960]

>>1471951
and a fuckton of flux

[Anonymous No.1471971]

2.4GHz antennas in 1206 format

>>1471949
no, clearly it's only 4/5 as good

[Anonymous No.1472069]

I posted in /qtddtot/ first after reading the OP here; they just sent me back.

I'm trying to learn small electronics, but I none of this stuff from Udemy is sinking in. Is there a course out there that really boils this down for dummies? I feel like the guy in my lectures rambles a lot and I can't really follow him at times.

[Anonymous No.1472075]

>>1472069
i picked up the basics from "practical electronics for inventors" and then built on that through dicking around in a simulator (multisim). both can be pirated.

[Anonymous No.1472091]

>>1472069
>they just sent me back
say thank you to the nice man who sent you here so you could get an answer in less than five minutes

[Anonymous No.1472099]

>>1472075
Why pirate multisim when LTSpice is free? Is multisim really better?

>>1472069
Look at the YT channels in the OP, if you ask politely someone here might post the file server/mega full of related e-textbooks. Having a look at simple circuits and figuring out how they work is a method that I find is pretty nice for getting a practical understanding of electronics, and BigClive takes apart tons of shitty $3 Chinese consumer electronics, reverse engineers their circuit diagram, and tells you how they work which can be great for this. Of course half the stuff just uses a microcontroller or is too simple to bother with, but there are a few 555 timer circuits to be seen, along with plenty of op-amps and comparators.

But you'll need to be more specific when it comes to what you want to learn if you want a better answer. Messing about in a circuit sim is also something I can recommend. Don't forget the ground node.

[Anonymous No.1472100]

>>1472099
in my opinion multisim's interface is marginally less cancerous.

[Anonymous No.1472101]

>>1472075
Thank you, I'll take a look into it.

>>1472091
Hey, I tried following the rules; it seemed more like a qtddtot issue to me.

>>1472099
I'll look into the channels. If someone did have a mega with a bunch of books, that would really help.

Long story short, my mom's boss wants to start a service for fixing medical lasers, and would like me to be his employee, where he sends me to clinics to fix the machines. He signed me up for Udemy but something about it just puts me to sleep. Like the guy seems really tedious and long winded at times, even though he's trying really hard to encompass everything.

[Anonymous No.1472115]

>>1472100
LTSpice on mac is so shit that there's only 3 buttons on the toolbar so you have to use shortcuts to do everything, so you may well be right. It also hasn't been updated in more than 2 years and keeps reminding you of that whenever you start it up. This is only the mac version though.

Medical lasers are pretty specialised beasts. If they're too new then there won't be much user-serviceability to them, but it's possible if you have the know-how. By fixing do you mean replacing parts with a screwdriver, replacing/fixing electronics with a soldering iron or hot-air station, or troubleshooting with an oscilloscope and logic analyser and reflashing ROMs and really getting down into the beast?

I've done a fair bit of laser physics so if it's on that end I can help also. If you haven't done this already I'd learn about op-amp topologies, transistor linear/saturation modes, comparators, diodes, capacitors, inductors, and filters/transfer functions and test them out in a sim. Learn about positive and negative feedback too.

[Anonymous No.1472140]

>>1472115
Speaking of LTSpice, why in god's name haven't they made a linux release?

[Anonymous No.1472148]

>>1472140
By the looks of things they're only focussing on the windows version. Isn't there some open-source spice/similar sim out there?

[Anonymous No.1472149]

>>1472148
I don't know, I run LTSpice with WINE. It's kind of clunky though.

[Anonymous No.1472168]

This fusing silicone tape is actually pretty good. Using it as strain relief on my fraying cables, it's much less rigid than heat-shrink (and doesn't have to be stretched over the USB-C connector) so doesn't just form another point where the cable is stressed. Stresses will occur at the original root of the cable because of how flexible the tape is (could remedy with more layers) but it doesn't matter nearly as much because the tape is atop it preventing the fray from being open to the elements. Note that the tape is not sticky at all so the cable can and will move underneath it if it isn't tight, so I wouldn't recommend it as a method of pull/tug-proofing the cable, just to stop it from continually bending at sharp angles at either end.

[Anonymous No.1472169]

>>1472168
Oh and it's $1.50 for 150cm: https://www.aliexpress.com/item/piss/32836970788.html
Which is a way better price than the name-brand "rescue tape".

[Anonymous No.1472171]

>>1472168

what's this got to do with anything? you were trying to reply to the ''Repairing extension cord for outside use'' thread? >>1471810

[Anonymous No.1472183]

>>1472148
the SPICE core, which takes a text file in and puts tables of numbers to be plotted, is midified BSD licensed. many commercial simulators are built atop this core

>>1472169
>epeen decoration store
10/10

>>1472171
nah, someone asked about strain reliefs a few threads ago and I suggested it. too lazy to provide link, gotta track all my incoming chinkparts

[Anonymous No.1472188]

>>1472183
>epeen
>silicone
Is this used for non-electronic means? [could it be?] I did think for a moment about using it to build up shapes, but it's not really rigid enough to be used like Sugru and its tape shape would make it fairly inconvenient.

I'll test to see if it sticks to other silicone objects.

[Anonymous No.1472221]

Just got a house, and because it was assembled by lazy Mexicans as everything is now some of the light switches are backwards where clicking them up turns them off and down is on. This is really triggering me and I want to fix it. Is it just as simple as swapping the live an neutral on the back of the switch? Is that even safe? or is the solution probably something more involved?


Side stupid questions, if the danger in working with electricity is it crossing your heart and stopping it, why don't sparkies have a grounding stap of sorts they strap to their working wrist or arm so it in case of shock it will only pass through their arm rather than hand to foot or hand to hand?

[Anonymous No.1472227]

>>1472221
>switches are backwards
Unscrew the 2 screws and flip the switch over

>why don't sparkies have a grounding stap
It's better to avoid being grounded at all. Also to work with one hand behind your back when working with live wires and there's a risk of getting shocked, like in a breaker box

[Anonymous No.1472238]

>>1472188
it provides a durable repair when attached to a leaking car radiator hose or home plumbing. it has even been used as an impromptu automotive accessory drive belt. that's how it got the nickname "rescue tape"
further discussion to /qtddtot/

[Anonymous No.1472245]

>>1472227
Most of the switches are gang switches and not every switch is wrong

Right, but in case you fuck or something out of your control happens

[Anonymous No.1472331]

>>1472148
There's spice and ng-spice. geda and kicad can interface to either of those.

The LTSpice simulator has been modified to handle switching transients better (I don't know if it's more accurate or just optimisation), as SMPS are a core area of LT's business. Also, it comes with models for most of LT's product line.

[Anonymous No.1472351]

Hi,
I recently modded my Switch to support Wireless Charging and it works like a charm.
I used this module: https://www.pollin.de/productdownloads/D351340D.PDF
I referenced this document: http://www.onsemi.com/pub/Collateral/AN-5086.pdf.pdf

The schematic looks like pic related.
Now I am kind of afraid of what happens if I connect a USB-C charger or the dock.
Do I need to add a diode to the +5V (and/or CC) line?

[Anonymous No.1472425]

>>1472331
What do those two spices have in the way of models?

[Anonymous No.1472503]

Something else occurred to me, /ohm/
You know how there's a bunch of comptetitively priced PCB prototyping companies out there and shit? like JLCPCB.
What about custom project box/enclosures?
Is there any place that can do that at a reasonable price? If my project has a metal case, I woul need bolt and screw holes machined, front panel needs square and circular holes cut out for displays and such. hole in the case to attach a crimped wire to for grounding etc. Doing all that on my own is a pain in the ass. Even finding a way to have it done is a pain in the ass
Not all of us have access to 6 axis CNCs

[Anonymous No.1472504]

>>1472503
as a meche my experience with CNC machine shops is that they have so much work right now that there's no incentive to price competitively. i tried to find someone local just to ream/bore a fucking through hole in a piece of bar stock for a personal project and couldn't do it.

[Anonymous No.1472519]

I'd like to control a somewhat large (3A-ish) DC current with a special switch rated too low (1A). This is a dumb hobby project, let's just leave it at that. I need the current to shut off immediately when the smaller switch goes off. I also don't really know much about /ohm/ shit, I watch a lot of eevblog but plenty of it is over my head.

I was thinking of taking my DC source which is sitting at around 18V and splicing a wire into a MOSFET; I put the smaller switch between my DC source and the MOSFET gate, when the switch is on the MOSFET starts conducting and when the switch is off the MOSFET stops too. Shouldn't have trouble pushing the current I need according to datasheets, and the voltage drop shouldn't be a problem.

My question is simply "will it work".

[Anonymous No.1472522]

>>1472504
I mostly made the 6 axis CNC comment out of exasperation
Point was mostly that I can't get anything even semi professionally made anywhere. I have to settle for janky set ups, or spend a bunch of money on tools and learn to do i tmyself. I'd get tools if I knew for sure i'd be doing it again on another box soon. or something

[Anonymous No.1472525]

>>1472519
pic related will work. note that most mosfets have a gate (control pin) voltage rating of 20V, so be sure your supply is actually a consistent 18V (or add a voltage divider).

[Anonymous No.1472530]

>>1472525
Remember to add a diode in reverse across the motor

[Anonymous No.1472535]

>>1472525
>>1472530
Thanks Anon!

[Anonymous No.1472538]

Going to try energy harvesting to trickle charge up a NiMH AAA cell with a few schottky diodes and long wires (floor to ceiling), any thoughts? I've heard that you can trickle charge a NiMH without worrying about overcharging or discharging, is this true? Or should I just use a supercapacitor (or even normal electrolytic capacitor) instead since it should be getting power almost constantly? I just plan on powering a simple 2-transistor NOR latch with it that toggles when my door is opened. Working on the circuit still.

[Anonymous No.1472539]

>>1471431
Rigols kinda suck, but they work very well and they'll do everything you could want as a hobbyist.
Even professionals use them, they might feel cheap but the bang for your buck is pretty good.

[Anonymous No.1472541]

>>1472539
Why are they bad?
But yeah anon get the one you had in pic related. It'll work a treat

[Anonymous No.1472556]

>>1472538
Here's the circuit, pretty janky. The left two transistors are the latch, while the right three are part of a delayed pulse generator, whereby the reset button is pushed, charges up the cap, triggering a transistor which is inverted, and both NOR'd by a third transistor which turns off for the short amount of time where both previous transistors are on. This makes a ~5s delay that will let me push the button before closing the door, since it's the door closing that will trigger the reed switch. I'll likely need a similar setup of delayed inputs being fed into a NOR gate on the reed switch side (probably with two reed switches) to ensure it only triggers on a closing, not on opening. It uses about 5µW, can I expect to get maybe twice this much from antenna energy harvesting?

[Anonymous No.1472561]

>>1472541
>Why are they bad?
Mostly the interface is slow and clunky, and some parts of it are poorly translated.
mine sometimes crashes when waving a waveform, I don't know if that's typical but they definitely lack some polish and care in the software.
the practical parts of the instrument work well and that's what counts, and at the price we're probably already getting more quality than we're paying for.

[Anonymous No.1472565]

>>1472561
Ah I understand. That does sound like a pain in the ass
>keysight oscilloscope
>costs over 2x as much
goddamn it

[Anonymous No.1472572]

DS212, anyone?
talk me out of it.

[Anonymous No.1472577]

>>1472556
Actually I might be better off throwing a counter IC in there instead of all the transistors. It will count each time the door has been opened and closed and is far less finicky. Since the current will be far more constant (no sudden spikes) I should be able to just use a low-capacity capacitor to power it. Though they do require a higher voltage.

I think I'll use a 4026, since it will give me a 7-seg output I can enable with a single button press, and it only uses 40nA quiescent. I'll be sure to put a big (10k) resistor on the display's common, which will also (probably) prevent me from needing display driving transistors.

[Anonymous No.1472592]

>>1472577
Also I won't need a hundred 1MΩ-to-10MΩ resistors.

[Anonymous No.1472593]

>>1472572
Low sample rate, limiting it's usefulness with modern electronics, but if you're ok with that...

[Anonymous No.1472607]

>>1472572
>not using a soviet-era dumpster scope

[Anonymous No.1472611]

>>1472607
I have a boat-anchor Tektronix 434 analog storage scope on the bench and a Leader LCD100 portable scope. I'm pretty happy with the Tek but desperately need an upgrade over the Leader.

[Anonymous No.1472632]

>>1472572
Buy a rigol or siglent DSO new, faggot

[Anonymous No.1472635]

>>1472632
sure I'll just carry one of those in my pocket.

[Anonymous No.1472649]

>>1472635
Any pocket multimeter is gonna be a piece of garbage. Get a real scope or don't waste your money. If you honest to god needed a portable scope (common in some telecom fields) there are solutions by many of the major manufacturers (eg. Keysight U1610A, Tektronix THS3000), both are expensive at $3300 and $4600 respectively for base models but I'm assuming if you are getting a portable scope you own or manage a telecom business and you are purchasing test equipment in which case you absolutely should not use a cheap-o pocket scope.

For home use it's highly unlikely you will need portability in a scope and even if you did the modern digital scopes from Rigol and Siglent aren't large or heavy and can be pretty easily moved with the only caveat being you need to plug them in.

[Anonymous No.1472656]

>>1472635
>>1472649
>Any pocket multimeter is gonna be a piece of garbage
I believe you mean scope, not multimeter, but yes this. If you want a portable scope I'd just get one of those $20 aliexpress ones, or the $15 ones and put your own case + battery on it, because otherwise you're wasting your money. I bought one of the $15 ones and it's fine for elementary troubleshooting and it's not really a money drain, but I'm not going to invest more than $50 in something that doesn't cut the mustard. Also TI, HP, and Hitachi are also probably good brands.

On whatever note that is, anyone know how easy it would be to run a digital desktop scope off a battery without an inverter/boost converter? The (somewhat less than modern) models I've seen don't have an external PSU to them, so surely they have some internal PSU making power rails and I was wondering if you could substitute this for one that runs off 12V or 24V for mobile troubleshooting out of a vehicle. But if not a 12/24V to 170/340V boost converter would likely do the trick since those PSUs rectify everything anyways.

[Anonymous No.1472657]

>>1472656
Don't be like this poorfag right here

[Anonymous No.1472658]

>>1472657
Hey I'm spending responsibly and I'll get a main scope when the needs outweigh the costs, but for now with a few audio signals and figuring out how a rotary encoder works, this thing is working just fine. But my shitty 9V wall wart is pumping too much noise into it when I look at the lowest scale, and there's also a few µV of DC offset in there somewhere.

[Anonymous No.1472661]

>>1472658
hey, fair enough
>spending responsibly
but if you buy that new scope, then you will never have to waste the 20 dollars on that shitty scope since you plan to eventually get the nicer one anyway

[Anonymous No.1472663]

>>1472661
Only $15, and I'll have you know that because of inflation this investment is actually paying off if I make $15 of interest from the bank within the time before I buy a better scope, assuming the better scope's price doesn't increase.

[Anonymous No.1472667]

>>1472663
You pay more if you buy sooner, if the price doesn't increase in the future though
spending responsibly my ass

[Anonymous No.1472670]

>>1472663
Is this your first bank account? Bank interest accrues at a ridiculously slow rate unless you have a lot of cash in the bank in which case it's still virtually nothing relative to how much you have in savings. Are you planning to use your $15 pocket scope for the next decade before you upgrade? Even then I'd argue you're better off just keeping the $15 in your account and having it accrue an additional $0.00001 of interest over that 10 year period. Now you have an additional $15.00001 you can put towards whatever $300 scope Rigol is offering in 2030.

[Anonymous No.1472681]

>>1471607
>That's some entry-level shit. Pic related I can solder by hand, and I'm not even particularly good.

I call bs unless you have 20/20 vision. lol

t. eyesight going to shit more and more every year who needs a digital microscope for micro soldering.

[Anonymous No.1472691]

>>1472670
Savings account interest is 0.1%, so if I had $1500 in a savings account I'd make up the $15 in a single year. It's been about that long since I bought the thing, though I'm not commenting on my financial situation.

[Anonymous No.1472703]

>>1472572
>tfw purchased an expensive scope and then used it twice
>tfw multimeter covers 99% of all my needs
don't do it, don't become me

[Anonymous No.1472711]

>>1472703
if multimeter covers 99% of your needs, your skills aren't improving

[Anonymous No.1472718]

>>1472711
scope is literally nothing more than a fast voltmeter

[Anonymous No.1472719]

>>1472718
voltmeter is literally a more useless oscilloscope

[Anonymous No.1472721]

>>1471126
I bought a red LED light from chyna but it doesn't meet my requirements in the following ways:
>not bright enough. I need about 10 times as bright.
>not the right shade of red, it's much too broad. I need something concentrated between 640 and 680 nm.

Supplying more power isn't an issue, but not breaking LEDs is.
Anyone know anything about buying LEDs to meet quite specific requirements?

[Anonymous No.1472725]

>>1472721
yes, diy

Also look into growing led lights, what you are looking is probably something like hyper red (660nm).

[Anonymous No.1472766]

>>1472721
LEDs tend to have a fairly broad emission spectrum, I'd recommend searching through digikey or one of them to see if they have a spectral width stat that you can sort by. There's also the option of using filters, but what do you want such a thin line for? A red laser diode with a lens (or several) could potentially work.

[Anonymous No.1472773]

>>1472649
>If you honest to god needed a portable scope
I troubleshoot/repair a wide variety of equipment across a few different buildings for the company I work for now. It'll be useful for car stuff too.
Having two channels plus signal generator output capable of PWM is great. the one thing I'd want to do that it seems like it can't do, is have it sit and collect data for hours.
I already have good benchtop scopes. this is not a replacement for those.

[Anonymous No.1472775]

>>1472656
>a 12/24V to 170/340V boost converter would likely do the trick since those PSUs rectify everything anyways.
Yes, this is the best option. Let it use the power supply that was designed to supply its rails.
Just make sure it doesn't have a transformer anywhere in there for line triggering or something.

[Anonymous No.1472806]

>>1472775
>12VDC to 170VDC boost converter
kek good luck finding one
Enjoy potentially fucking up your electronics supplying DC to an SMPS rectification circuit

[Anonymous No.1472819]

>>1472806
>Enjoy potentially fucking up your electronics supplying DC to an SMPS rectification circuit

ahah, what a retarded comment. the rectifier will handle a voltage of either polarity. giving it just one polarity just means that half the circuit is idle.

[Anonymous No.1472829]

>>1472819
which could matter if the input circuit uses a C-W voltage doubler on the line side to get 340Vdc (not uncommon)

[Anonymous No.1472883]

>>1472829
Or if it has active PFC. Or a choke for inrush current limiting.

[Anonymous No.1472906]

>>1471129
>>1471130

VLSI design you'll need a masters degree to find work in the field due to how specialized it is, computer engineering can find work with just a bachelor's. There's plenty of jobs for both though if that's what you're asking

[Anonymous No.1472918]

>>1472906
>VLSI design you'll need a masters degree to find work in the field due to how specialized it is
Why's that? I wanted to do embedded programming, and embedded system design

what does the day of the average computer engineer look like? What do they do? I know it's a mix of both computer science and electrical engineering, but what jobs do they usually get hired to do?

[Anonymous No.1472933]

>>1472918
Embedded programming's a little different from system design. The reason is that courses that teach VLSI are usually only offered in 4th year of a program due to the background knowledge needed, and there's a lot of material and new technologies to cover so undergrad courses can't cover it all. Of course it's possible to learn on your own using textbooks or online, etc but convincing employers to hire you could be more tricky. You would also likely have to learn cad and simulation software.

[Anonymous No.1472946]

>>1472883
neither of those would disagree with a DC input, unless the PFC controller has some auxiliary features that get confused. the traditional PFC control loop should just act like a boost converter with a DC input.

[Anonymous No.1472952]

>>1472883
Or it relies on a tapping an inductor to accumulate an initial charge on a capacitor to power control circuits
Active PFC is present on every SMPS though, which is common to about 99% of all modern electronics

[Anonymous No.1472974]

>in canada
>want to buy some capacitors
>digikey.ca
>order some shit
>disclaimer about export fees and duties
>20 dollar shipping to top it all off
what the fuck

[Anonymous No.1472980]

>>1472952
>Active PFC is present on every SMPS though

I mean, if it's an ATX power supply, yeah. But those Chinese PSUs that are omnipresent on eBay and AliExpress...not so much.

[Anonymous No.1472982]

>>1472974
>>>/leaf/

[Anonymous No.1472984]

>>1472952
> Active PFC is present on every SMPS though, which is common to about 99% of all modern electronics
Not even close. On consumer gear, you normally only find it on computer PSUs, and then only on the better ones.

It would probably be more common if the EU hadn't caved to the PC industry on EN61000-3-2. The original limits would have effectively required active PFC for anything over ~500W.

[Anonymous No.1472986]

>>1472980
>chinese shit
Yeah that's true
I suppose some lower power shit don't have active pfc and just have bulk caps doing passive pfc

[Anonymous No.1472988]

>>1472974
FOR YOU
http://www.island.net/~kdbrown/WhereToBuy-Canada.php

[Anonymous No.1472997]

>>1472775
Is line triggering likely to be used in a scope? I suppose it could be a worthwhile feature on

>>1472829
>C-W voltage doubler
I doubt it, those things are for very low power, and I can't think of anything that a digital scope would use such high voltage for. If it was analogue then it probably would use a transformer.

>>1472952
If you hacked around inside the PSU chances are you could bypass the PFC (and rectifiers) and send DC straight to the capacitors. Using the existing DC from before the boost converter to more efficiently power the step-down converter itself is also a possibility, provided that's not one too many references to ground. But usually even if they are powering it from a coil it will be a feedback winding in the main flyback transformer.

I wonder what the minimum running voltage of those buck converters is? 140VDC (since those devices often say 100-250VAC on them)? Do the old scopes with voltage switches on them use multi-tap transformers?

[Anonymous No.1473001]

how many amps do i need to plate a coyote skull?

[Anonymous No.1473009]

>>1472988
Thanks anon!
Checking some of the "canadian sources"
They seem to have janky websites and/or poor selection
Was looking for some rubycon or nichicon caps with a high rated life for recapping a speaker control board. They're shity speakers and the board gets hot
Also some excess caps for tinkering more with electronics later on

[Anonymous No.1473011]

>>1472997
>If you hacked around inside the PSU
Yes, but that anon is talking about straight connecting 170VDC to where the usual 120VAC line would go. Not hacking around
Not to mention finding a 12 to 170V boost converter that has the appropriate wattage and protections
And even then he has too many what ifs to deal with
it's a terrible idea and he should feel terrible
If you've measured the bulk caps on the inside of power supplies (i have) they sit between 160 - 170 VDC

[Anonymous No.1473014]

>>1473011
But I am that anon. AFAIK digital scopes don't use that much power, so finding an adjustable DC-DC converter wouldn't even be too hard. Making my own converter is also an option, provided I start with a good converter IC and follow PSU designing guidelines.

[Anonymous No.1473025]

>>1473009
maybe try leaf Newark or leaf Arrow or leaf Allied. you probably won't find a full-line distributor. you might (or might not) do better chinking it

[Anonymous No.1473027]

>>1473025
I bought some shit from leaf arrow last time i ordered a bunch of stuff
it was almost like he said - saddled with high shipping, and a shitton of customs fees and shit. added another 30 - 40 dollars to my order iirc. had to pay at the door and everything
I suspect digikey would have done that to me with their 20 dollar shipping (unless i buy 100 dollars of shit) and their disclaimer that i may get charged duties etc

[Anonymous No.1473034]

Ok looks like my rectenna won't work without a tuning circuit. Are AM frequencies the ones with the most energy?

[Anonymous No.1473035]

>>1473034
you'd think some asshole with antennas and a spectrum analyzer would go measure this, but i'm actually having a hard time finding any info with google.

[Anonymous No.1473037]

>>1472997
Voltage doubler used to be common to get 120/240 volt support cheap, now kinda rare using boost converter PFC instead for full range.

DS1054Z can run down to about 40V DC on the line input.

[Anonymous No.1473039]

>>1473035
Me too. If I make an upconverter for my $20 SDR (and can characterise an antenna's frequency response curve) I'll be able to analyse all the lower frequencies, but for now I'm shooting in the dark. At least I've got some trimmer caps lying about, but I'll have to wind a coil I suppose.

[Anonymous No.1473040]

>>1473039
Broken screen? No problem.

[Anonymous No.1473052]

>>1473040
i want VFD screens :(

[Anonymous No.1473059]

>>1473052

[Anonymous No.1473068]

>>1473059
yes exactly

[Anonymous No.1473070]

>>1473068
I prefer the nixie aesthetic though, especially ones with an inverted 2 as the 5. Plus multiplexing the multi-digit (let alone multi-dot) displays on a VFD means you get some flicker when you film it. Not as much of a pain to drive as EL displays though.

[Anonymous No.1473080]

>>1472997
>I wonder what the minimum running voltage of those buck converters is?
A few phone chargers I tried would work down to 75-80 volts DC. I never tried anything more complicated or higher power.

[Anonymous No.1473103]

>>1473070
Nixie's are cool here and there
But they're tits expensive
VFDs are like some odd mix of retro and future looking tech that i like, on top of oleds or whatever
7 segment is out these days though

[Anonymous No.1473114]

>>1473103
Nixies aren't that expensive if you go for the smaller ones, and stereos were using VFDs up until the 90s, so they're not exactly retro.

But this EL display is where it's at.

[Anonymous No.1473121]

>>1473103
IN-14 + IN-19 is the combo I'm looking at, $5 each for the 1N-19 symbol tubes and $15 each for the IN-14 numerical tubes.

If that's too expensive, try the IN-1 paired with the IN-7, which are only $2.13 and $2.84 respectively, over at tubes-store.com.

Pic related is a spreadsheet I made, where the colour groups on the side represent tubes that could go together. "•" is for single dot tubes, "." is for decimal points, "—" is for bar graph tubes, the rest should be self explanatory. Tubes-store only has IN-series nixies (plus some far more expensive ones) so I only covered them, but there's also a lot of other ones available on eBay, and here's a library of all sorts of nixies: http://www.tube-tester.com/sites/nixie/trade01-nixie-tubes.htm
Guy has all sorts of other shit on his site too, like decatrons and such.

[Anonymous No.1473126]

>>1473103
I'm into these chip-on-board style LED displays with magnifying lenses.
DL1416T

[Anonymous No.1473130]

>>1473114
I don't care about retro. I just like their aesthetics

[Anonymous No.1473132]

>>1473126
Aren't those oldschool as fuck? I vaguely remember seeing those in some 1970s calculator. Pretty neat aesthetic

>>1473121
some czech guy is making nixie tubes and they cost like 200 dollars each

[Anonymous No.1473139]

>>1473130
Fair enough. My main beef with them is more that you could buy very similar looking LED 7-segment displays. For VFDs with other shapes, like bar graphs, VU meters, dot-matrices, etc. they're certainly pretty cool looking. But I can't really think what to use one for. There are some RGB VFD dot matrices (IGG4 64/64M2) on eBay, which look fucking awful to drive but they could be a fun project I guess. Oddly enough there's also some less than common displays on aliexpress (https://www.aliexpress.com/item/D/32638805488.html)), but again I can't really think of any good uses for them.

[Anonymous No.1473145]

>>1473139
>similar looking LED 7-segment displays
True enough. I'd like a colour similar to that of VFD or even the same. Rarely see blue though. It's usually red or green that i've ever encountered
One thing I don't like about vfds is just the fact that you have to drive them with high voltage

[Anonymous No.1473153]

>>1473145
I wonder if you can get a 7-seg display with 3 cathodes and 7 anodes, each of the cathodes being for R,G,B so you can PWM them for any colour you want. From what I've seen you can get RGB ones with 25 pins on them and ones with 3 pins (I2C I'm guessing). But there doesn't appear to be a proper cyan 7-seg display out there. But it looks like the 25 pin ones can be soldered into to the 7-3 layout I described earlier (plus one decimal point pin), which is fine to do if you want all the segments to have the same colour.

[Anonymous No.1473157]

Why do the selinoids in pinball machines have giant resistors bridging them? What's the point?

[Anonymous No.1473158]

>>1473153
Wouldn't you just make cyan by rgb'ing it?

[Anonymous No.1473159]

>>1473158
Yes that's what I was getting at, but it does mean you have to buy a specialised model of a 7-seg display, and personally I don't like the look of its segments.

[Anonymous No.1473166]

>>1473159
the lighting isn't quite uniform enough to give that VFD look

[Anonymous No.1473184]

>>1473157
Flyback suppression. Without it, the inductance will cause the switches to arc.

[Anonymous No.1473201]

routing discrete logic is fun

[Anonymous No.1473204]

>>1473159
Why not just get an OLED display and paint 7-segment graphics on it?

[Anonymous No.1473206]

>>1473204
Why bother with a 7-segment style display if you're using an OLED anyhow? 7-segs are cheap and easy to drive, an OLED is certainly not the latter, and an RGB OLED isn't the former either.

[Anonymous No.1473209]

>>1473201
>not moving the DIP-14s by half a pitch so the packages line up with the DIP-16s

[Anonymous No.1473210]

>>1473209
my autism prefers having the ends in line rather than the centers

[Anonymous No.1473212]

>>1473206
>7-segs are cheap and easy to drive
>I can't get the one I want
QED

[Anonymous No.1473216]

>>1473210
but the packages themselves are the same length

[Anonymous No.1473217]

>>1473216
don't you understand that the holes have to be on a perfect grid or something terrible will happen

[Anonymous No.1473267]

How do i pick best resistor values for a voltage divider?

Basically 1ohm and 2ohm resistors give me same voltage split and 100k and 200k resistors

[Anonymous No.1473270]

>>1473267
Not low.
1ohm + 2ohm wastes a lot of power.

[Anonymous No.1473275]

>>1473267
Anything you attach to a voltage divider will affect it, say you have a 1kΩ load that you want to put 3V across but have 5V at your disposal. If you use a divider made of a 20kΩ resistor and a 30kΩ resistor then the 1kΩ load will be in parallel with the 30kΩ resistor, making the effective combined resistance equal to 968Ω, hence the divider will have a voltage of 0.231V instead of 3V.

Do this with a divider of a 200Ω and a 300Ω resistor instead, and you'll have an effective combined resistance of 231Ω, giving you a voltage at the middle of 2.68V, which is much closer to the 3V you wanted. So it's a balancing act between wasting power and accuracy. Typically we don't use voltage dividers to power things because their output voltage varies with current, so instead we use linear regulators, or buffer the signal. Many ICs have an input impedance of many millions of ohms specifically so that they don't cause the outputs of other components to be effected, which in this analogy would mean using a ~10MΩ resistor instead of a 1kΩ resistor.

[Anonymous No.1473277]

>>1473267
As high as practical to keep power loss down as anon said

It depends on the load you're going to put on it. If it varies, the divider should have a lot less resistance than the load.
If you were to put a 100K load on a divider in the 1K range, It wouldn't change the divider voltage much much because 100K barely effects the circuit.
If you were to put a 1K load on a divider in the 100K range, It would throw the votlage way off because the 1K becomes part of the divider and effects it a lot

If the load is constant, Just make it part of the divider, by adding a series resistor.

[Anonymous No.1473280]

>>1473267
The three anons above me already mentioned why you don't typically use low values because of wasted power but I'll add that using very high values can also be bad depending on the circumstance. For instance if that divider is acting as biasing for some other circuitry and it needs to deliver some minimum amount of current in order for the circuit to work properly. In addition using very large value resistors creates more Johnson–Nyquist noise which increases in proportion to resistance. In general using dividers consisting of resistors between 1k and 100k given low-medium supply voltages from 0-30V will usually give you good results keeping power dissipation acceptable and giving good noise performance.

[Anonymous No.1473290]

>>1473280
Isn't there a problem when the voltage divider resistance is similar to an op amp's input impedance?

[Anonymous No.1473298]

>>1473290
Yeah. With voltage amplifiers you generally want low output impedances feeding into high input impedances because the voltage divider created by the output and input impedance is very small and barely divides the voltage at all. The greater the difference the better. When the output and input impedances get close or the output is greater than the input it begins to load the amplifier down, dividing down your input signal and reducing your gain. This isn't a major issue for FET input op-amps since their input impedance is on the order of hundreds of gigaohms to teraohms and good luck getting resistors even close to that, you can maybe find 1Gohm if you wanna spend a lot of money. For older op-amps with input impedance in the megaohm range it's worth remembering though.

[Anonymous No.1473305]

i accidentally made a solder poopoo over thwo tiny contacts and now they are fused together, i have some John Wick, but i never used it before, how do i remove solder with him/it?

[Anonymous No.1473311]

What's the formula to figure out the time it takes to charge a capacitor if you are using a constant current source to charge it?

Like, an active constant current source, eg. current mirror, not a simple resistor.

[Anonymous No.1473313]

>>1473311
A farad is a columb per volt. An amp is a columb per second.
If you charge a 1 farad capacitor with a current of 1 amp, it will charge at 1 volt per second. and so on.

[Anonymous No.1473314]

>>1473132
they are pretty oldschool. they were pretty advanced for their time, too. 64 character alphanumeric display, all the decoder/driver circuitry is built right in. the timing is so forgiving you can even use them manually with switches.
very arduino-friendly. I got a big tube of them by chance and I use them wherever I can now.

[Anonymous No.1473318]

>>1473314
Damn they sound comfy
I know i said I don't like things for simply being retro, but the retro aesthetic is definitely comfy

On another note, I went down a rabbit hole of soldering videos from that androkavo guy
Started looking up soldering stations
>mfw they can be hundreds, or even thousands of dollars
what the fuck its a soldering station
I know they're marketed to companies and production houses but still. jesus christ
>hear about one that detects heat load and automatically increases power to try to get back to temperature
or some shit like that
>look it up
>costs like 700 bucks or some shit
what the fuck

[Anonymous No.1473328]

How much current does the coil on one of these pull? Can I operate one straight from a microcontroller or would I need a transistor or something?

[Anonymous No.1473340]

>>1473328

read what it says on the side, or the datasheet. it varies with size, from maybe 20mA to 20A. a typical size will need around 100-300mA which is transistor territory. dont forget to put a reversed diode across it.

>>1473305
>i have some John Wick, but i never used it before,

forget your John Wick, try some Aeon Flux.

[Anonymous No.1473356]

>>1473184
Ooh. So would running a selinoid without one for a long time damage it? Or just be dangerous?

[Anonymous No.1473359]

>>1473318
oh they are comfy. it's hard to get a good picture of it but the lines are so crisp and bright it's a very unique look.

soldering stations can get expensive but you don't need to spend a lot of money on one. The number one requirement for getting good results when soldering is temperature control. All temperature controlled irons will "detect heat load" because they apply more power when the tip temperature drops.
Mostly I use Weller WESD51 stations which go for under $150. I do tiny SMT rework stuff as well as big things. For myself I use a TS100 mostly for through-hole and have no complaints.

Your success in tiny precision soldering doesn't depend so much on your iron as it does on your ability to see. You don't even need a super fine tip as long as you can position it accurately.
Save on the soldering station, spend on a magnifier.

[Anonymous No.1473391]

>>1473356
arcing on the switch will create pitting, decreasing the contact area. eventually the switch will either fail open or will weld itself closed. how fast that happens depends entirely on how robust the switch is.

if you're missing a resistor, you can just replace it with a reverse biased diode.

[Anonymous No.1473418]

>>1473391
> if you're missing a resistor, you can just replace it with a reverse biased diode.
That only works for DC. I thought pinball tables used AC for the solenoids; the early (electro-mechanical) ones used AC for everything, but given how much electronics are in a modern one they might just use DC throughout.

Also, a diode allows the flyback current to circulate freely, so it can cause the solenoid to remain active for a while after power has been removed.

[Anonymous No.1473419]

>>1473418
>a diode allows the flyback current to circulate freely, so it can cause the solenoid to remain active for a while after power has been removed.
how long is "a while" for you? The freewheeling current is very rapidly burned off in the windings' own resistance.

[Anonymous No.1473420]

>>1473418
you're right about the AC, my mistake.

[Anonymous No.1473427]

>>1473114

[Anonymous No.1473433]

>>1473419
It's long enough that there are DC applications where you don't want to use a diode. E.g. the solenoids in electro-mechanical printers typically use R-C or R-C-D snubbers rather than a diode. It can be an issue for relays because a slowly-decaying field reduces the speed at which the contacts separate, increasing arcing.

[Anonymous No.1473441]

>>1473318
it's more than one soldering tool in one convenient case. it can be nice not having a shit-ton of boxes on your bench
also what next anon says. don't try to cheap out with the $30 USB "microscopes". the only thing good to say about them is that they are better than the naked eye, sometimes, for some things. don't be like me

>>1473359
I have ~$150 budget for magnification. considering the Andonstar ADSM201 but probably won't do enough still/video to justify an HDMI scope if something optical can be gotten for cheaper. recommendations?

[Anonymous No.1473442]

>>1473114

[Anonymous No.1473444]

hello /ohm/

i built a led chaser circuit with 555 timer and 4017 decade counter. now, i want to hack it so the leds will fade in-out instead of on-off blinking. i'm currently reading up on RC circuits, but i'm a bit slow in the head so... thought i'd ask you if something like pic rel would do the trick.

i'm terribly sorry if this is idiotic and belongs in the stupid questions thread

[Anonymous No.1473449]

>>1473441
I've used Amscope stuff, but it's expensive for hobby use. I'm not familiar with anything cheaper.
I can say though that having stereo vision makes a huge difference. I've worked in mono before (camera in the other eyepiece) and it's way harder to see where your iron and solder are. Try and find something stereoscopic.

[Anonymous No.1473451]

>>1473444
Try simulating it. Download LTSpice, make that circuit, and set up a transient simulation.

[Anonymous No.1473452]

>>1473444
it doesn't really belong in any thread.
Build it and see if it works. If it doesn't work and you can't figure out why, then come back and ask for help.

[Anonymous No.1473455]

>>1473451
>LTSpice
i checked out the download page. apparently not available for linux. i'll try to find an analogue tho
>>1473452
i'm gonna take a long trip to the nearest electronics shop tomorrow to stock up on passive components and experiment. just thought i'd gather some comment from more experienced people beforehand to make the trip worthwhile. maybe the circuit in imagined is total bullshit haha

[Anonymous No.1473459]

>>1473455
LTspice will run OK with Wine.

[Anonymous No.1473462]

>>1473441
>>1473359
Oh I have a cheapy chinese knock off already. I was just curious what's out there
Can't believe some of these are so fucking expensive kek

[Anonymous No.1473463]

>>1473459
thanks!

[Anonymous No.1473466]

>>1473463
Intuitively, though, it should work. If you want to pick component values to get certain performance, that's going to be harder and require some math.

[Anonymous No.1473468]

>>1473466
yeah i have that feeling too, but countless times my feelings have been wrong. i'm basing my experiment on the premise that i need only to keep the current limited through the second resistor during a "low" in the pulse so i don't blow the led during discharge. i'm wondering tho what will happen to the led brightness while the capacitor is charging. i'm excited to find out

[Anonymous No.1473479]

>>1473418
This is an old electro mechanical one. So I'll try and order another gigantic resistor and solder it in

[Anonymous No.1473509]

I'm looking for a way to drive a small low-voltage light bulb from a 12 V battery pack with high efficiency. I found this circuit that uses an LM2575 switching regulator to deliver a constant current into a load.

Is there any way to control the lamp current with cmos logic (or microcontroller i/o) pins? I'd like to (at least) be able to control the lamp current between two settings, namely: bright light for normal operation and power saving.

[Anonymous No.1473518]

>>1473442
Who did that?

[Anonymous No.1473527]

>>1473444
We had this question with lighthouse anon a few threads ago could find it if you sift through the archives, it was definitely less than a month ago.

From what I remember, you put a capacitor across each LED but have the LED's dropper resistor before the capacitor, so it takes time to charge the capacitor up through the resistor and down through the LED. But depending on the voltage you may also want to put a resistor in series with the LED and/or the capacitor to get the charge/discharge times to be equal. Best to do that tuning IRL since the sims don't really give a good impression of the light output of an LED.

>>1473509
If you want to use a microcontroller then it's probably simpler to just use the MCU with a MOSFET as the switching controller: just PWM the inductor, have a current shunt feeding into the ADC, and use the shunt voltage to figure out whether the duty cycle should be raised or lowered. Even simpler, just turn it off when the current reaches the threshold and turn it on again when it drops, in case you're not able to program the former method. But is there any reason you can't just use a potentiometer to control the current? Doing that would be a simple matter of replacing the microcontroller in this diagram with a comparator and a 555, plus a handful of passives.

[Anonymous No.1473537]

>>1473518
OC

>>1473509
add a resistor in series with a MOSFET, all in parallel with R3. select MOSFET for appropriate turn-on voltage according to the logic level you'd be using with it
or, there are constant-current LED driver ICs that accept PWM dimming inputs, which you can probably adapt to drive a lamp instead
or, with some regulators, you can abuse an external soft-start pin to set a voltage (ergo current) limit, like so http://www.ti.com/lit/an/slva451a/slva451a.pdf (just replace the LED with your lamp of choice)

[Anonymous No.1473538]

>>1473509
example constant-current LED driver IC that accepts PWM input https://www.onsemi.com/pub/Collateral/CAT4201-D.PDF

[Anonymous No.1473541]

>>1473527
>you put a capacitor across each LED but have the LED's dropper resistor before the capacitor
you shouldn't do it that way. then the capacitor voltage will only be exactly equal to the LED's forward voltage, and as soon as it drops slightly, the LED will go out.

[Anonymous No.1473565]

>>1473541
A diode's forward voltage isn't some on-off transition point like a MOSFET's threshold voltage, but rather an exponential curve. Put 1mA then 20mA through an LED, there will be a noticeable difference if you measure the voltage.

[Anonymous No.1473567]

>>1473565
oh not to mention that as the LED draws less current as the voltage across the capacitor goes down, the capacitor will be losing charge slower making the ramp off even smoother.

[Anonymous No.1473568]

>>1473565
>on-off transition point like a MOSFET's threshold voltage
MOSFETs have a linear region

[Anonymous No.1473576]

>>1473568
Yes and it's about a µV across

[Anonymous No.1473580]

>>1473576
several hundred millivolts or more, depending on the max current you're pulling through it

[Anonymous No.1473581]

>>1473580
Woah that's way more than I expected. I've never managed to get a FET in the linear region without any feedback, so I guess I wasn't trying hard enough.

[Anonymous No.1473584]

>>1473444

this was answered a couple of threads back. >>1465578

[Anonymous No.1473588]

yeah that (deleted) image doesn't really apply to an LED because muh non-ohmic

[Anonymous No.1473589]

>>1473444

this was answered a couple of threads back. >>1465578
>>1465652

[Anonymous No.1473590]

>>1473588
>muh non-ohmic

and whose eyes can tell the difference?

[Anonymous No.1473591]

>>1473590
you and me both, baby

[Anonymous No.1473592]

>>1473591
blue is the voltage across the LED

[Anonymous No.1473597]

>>1473592
>220µA
you're seeing something? what is that, a white LED?

[Anonymous No.1473599]

>>1473597
Well I had to turn the resistance way up otherwise the cap would discharge too quickly. Besides, 200µA is definitely visible.

[Anonymous No.1473601]

>>1473597
Not him but it's not about noticing but above precise control
He could sinusoidally control the pulse width to get a perfect gradually dimming and brightening pulsing effect. whereas if you tried just directly driving with a sinusoidal wave, it'd just blink because of the non linear nature of LEDs

[Anonymous No.1473607]

>>1473599
>discharge too quickly
>not putting a diode in series with the resistor
>not adding a resistor after the cap as well

>>1473601
that's actually not wrong. lighthouse revolving beacons are focused into a tight beam for maximum relative brightness over distance. from a fixed point, one would see a bright flash with sloped edges
https://www.youtube.com/watch?v=2X7saQT0BuQ

[Anonymous No.1473615]

>>1473607
>>1473601
And pwm allows you to make the LED behave arbitrarily, s long as you have the proper waveform with which to modulate the PWM pulse width

[Anonymous No.1473623]

>>1473615
certainly, imho PWM would be far preferable for this application and much more direct, and also cheaper, but iirc anon wanted to do it analog style for some reason, so here we are

[Anonymous No.1473631]

>>1473623
Ah, fair enough
I empathize, but anon should definitely look into pwm anyway. its a good learning experience and another tool he can have under his belt for his projects

[Anonymous No.1473650]

>>1473615
PWM means multiplexing a signal at high frequency, which is kinda a pain but not that awful. Say he wants 20 brightness steps, with maybe 8 LEDs, with a frequency that isn't noticeable to the human eye like 60Hz, so that's 9.6kHz or 9.6kb/s. Give or take a factor of 2. Then the latch input needs to be toggled once every 8 cycles, 1.2kHz.

[Anonymous No.1473654]

>>1473623
>iirc anon wanted to do it analog style for some reason

he's got 10 LEDs so, being a sane person, he doesnt wanna wire up 10 externally-controlled PWM circuits, then create some logic that's gonna vary the the intensity of each one in sequence. a sane person will wire up some caps which substantially accomplish the same effect.

[Anonymous No.1473658]

>>1473650
1.2kHz ain't shit. for an 8 second revolution time at 60fps you would need a LUT that's 10*48 = 480 bytes long. just play that back 10 times, advancing the start LED once for each round. plenty of time to simulate PWM in code
also see Pic related, $3, which would need to be updated only 60x/sec or however fast the animation needs to run. I see no reason this couldn't be driven by an ATtiny85 let alone an ATmega328

>>1473654
what's insane about Pic related?

[Anonymous No.1473661]

>>1472140
>why in god's name haven't they made a linux release?
Lol why bother when it runs awesome in wine?

[Anonymous No.1473663]

>>1473654
>10 externally-controlled PWM circuits
shift registers my bro, he's already using one of them for the existing 555 circuit.

>>1473658
I wasn't saying 10kHz was particularly an issue, just that it would be a bit of a pain to work with, mainly for troubleshooting purposes. And that PCB looks a little overkill compared to a single shift register and bit-banging the fuck out of it, if you can be bothered programming it.

[Anonymous No.1473668]

>>1473658
>what's insane about Pic related?

- a computer chip to flash some LEDs is overkill by a factor of 1000 minimum
- requires programming, which means hundreds of hours of a learning curve for a noob. and it's not just programming, but learning the IDE, the tools, the cables, the intricacies of the software switches. 1000 details to memorize, any one of which can derail you for hours.
- may require SMD soldering and the attendant tools.

[Anonymous No.1473672]

>>1473663
>shift registers my bro,

no good. he needs to have various states for each LED, say 5 off, 2 at 33%, 2 at 66% and one at 100%

[Anonymous No.1473673]

>>1473663
>shift registers my bro,

no good. he needs to have various states for each LED, say 5 off, 2 at 33%, 2 at 66% and one at 100%. and these states are rotating.

[Anonymous No.1473674]

>>1472140
it's not like they're making any money from it, directly

>>1473668
>overkill
depends how you're scoring your golf game. as to cost, space, and flexibility, the micro route wins easily
>programming
that is true. I keep forgetting this isn't a skill that kids learn in fifth grade anymore
>SMD
not particularly, esp if there's only one of those boards on the bus. if he instead uses something like an arduino nano, surely there's 10 outputs for soft PWM and the board wouldn't be needed. the rotating states aren't a problem, depending on how many brightness levels are required

[Anonymous No.1473689]

>>1473650
pwm is dick easy and you're making mountains out of molehills
i managed to make a simple pwm assembly with some op amps, caps, and varaible resistor.
just substitue varable resistor with arbitrary waveform input. yikes buddy

[Anonymous No.1473712]

>>1473673
yes, just have a latching shift register and set it in cycles of maybe 20 pulses per cycle:

In arrays of LED numbered 1,2,3,4 for a 4-bit register (4 LEDs) with 8 pulses per cycle, this gives duty cycles of 100%, 50%, 25%, and 12.5% respectively:
>1,0,0,0
>1,1,0,0
>1,0,0,0
>1,1,1,0
>1,0,0,0
>1,1,0,0
>1,0,0,0
>1,1,1,1
All fed into the data input of the shift register sequentially, latching after each line. You just have to feed it in fast enough to not be visible, which isn't difficult with modern hardware. This cycle would probably be repeated ~10 times (perhaps with finer changes in duty cycle) before it moves on to the next cycle:
>0,1,0,0
>0,1,1,0
>0,1,0,0
>0,1,1,1
>0,1,0,0
>0,1,1,0
>0,1,0,0
>1,1,1,1

>>1473689
>t. didn't read the question
It's for 10 LEDs which each have a different brightness as varies with time, and he wants to control it with an MCU, which I'm assuming doesn't have enough analog outputs. 10 Digital pots could work through I2C, but it's a bit clunky compared to a single register.

[Anonymous No.1473718]

>>1473712
adafruit once had a surprisingly good idea for bit-banging PWM: output the bits according to the bit weight, and then wait for a time according to the bit weight
>peel off all your 1 bits, output, wait 1 time
>peel off all your 2 bits, output, wait 2 times
>peel off all your 4 bits, output, wait 4 times
and so on
also, if there's a micro involved and he can bit-bang, who needs a shift reg?

[Anonymous No.1473722]

>>1473718
Shift register because he probably doesn't want to waste 10 PWM pins on his micro. But if he does then more power to him.

[Anonymous No.1473724]

>>1473722
there won't be 10 PWM pins on most micros (except maybe bluepill), so he'll be bit-banging something on a timed basis anyway

[Anonymous No.1473735]

>>1473724
Oh yeah, not even the mega328p.

[Anonymous No.1473737]

>>1473724
You don't need a PWM pin for this. This low of a frequency is easily doable in software.

[Anonymous No.1473747]

>>1473737
Not him but
software is easy, and flexible
But doing it in hardware is so much more fun, and rewarding. At least for the first like 10 times that you do it kek

[Anonymous No.1473750]

>>1473737
pin1 = high
wait(1)
pin2 = high
wait(1)
pin3 = high
wait(2)
pin4 = high
wait(2)
pin1 = low

Sounds fucking awful, but what do I know about real-time MCU programs. I'm guessing you can make a function to handle the PWMs of all the pins turning on and off at the right times, but either you make it calculate the right delay to add between each pin command AND know how to reorder the pin commands, or just have a bunch of single, smallest delays and a whole lot of IF clauses between each. Either way, it's hardly elegant.

[Anonymous No.1473763]

>>1473750
with lolduino using software PWM is ezpz, analogWrite(pin number, duty cycle out of 1024) the high/low/wait is handled automatically.

[Anonymous No.1473767]

How do I into big boi micros instead of arduinos?

[Anonymous No.1473768]

>>1473763
Yes but that's only for the 5 PWM pins, and this guy needs 10. This problem was already clarified here: >>1473724, hence the reason behind my post.

[Anonymous No.1473771]

>>1473768
depends on the particular chip, ESP32 with lolduino framework gives you 16 independent semi-hardware PWM channels.

[Anonymous No.1473784]

>>1473771
Those ESP32s are actually really good, apparently they've got a really good ADC too. A bit too IOT flavoured for my purposes though, and just generally overbuilt. Recently I've been going low-scale with a few DIP ATtiny85s, but is there a good lightweight MCU that's cheaper? SMDs are a little out of the question since I'd just slap them on a breakout board anyways. Even so, it looks like the STM8s are like 50c each if you buy in bulk, similar prices for some DIP-8 PICs. Anyone had experience with either?

[Anonymous No.1473790]

What the fuck? They make modules for shifting logic singals between 3v and 5v? why the fuck nobody told me that? why the fuck did i made all those fucking voltage dividers?
fuck you

[Anonymous No.1473791]

>>1473790
Dividers are just fine for step-downs, but for step-ups you need something better. One or two transistors is probably fine, though if it's a two-way line I suppose you'd have to use some specialised IC, I can't even imagine trying to create such a circuit with op-amps. I think you'd need current feedback of some sort...

[Anonymous No.1473792]

>>1473791
i just ordered 20 bidirectional 4 channel shifters that can go between 3 - 5 and 5 - 3 and even shit like 1.2 etc and just for a buck a piece, i fucking love chinks
All i have to do is skip ONE lunch and i can literally buy enough electronics from chinks to build a robot

[Anonymous No.1473798]

>>1473792
Fuck, don't remind me of my hunger pains. Or it might be food poisoning, that chicken was sitting uncovered for a while in the fridge.

[Anonymous No.1473799]

>>1473798
tfw severe diarrhea running to toilet ever 30 mins or so

[Anonymous No.1473820]

Is there any difference between "SCK" and "SCL"?
I have one I2C oled screen with SCK and one with SCL, trying to troubleshoot why one of them doesn't work properly. The image is squished into the top 1/4th of the screen and the rest is snow.

[Anonymous No.1473823]

>>1473799
>30 mins
>severe

also if you eat like a pig, expect your body to treat you like one

[Anonymous No.1473828]

>>1473823
I don't
But occasionally I eat something bad or something and that shit passes through me. Or in worst cases, constant diarrhea
What now you pig fucker?

[Anonymous No.1473831]

>>1473820
Looks like it's not compatible with ssd1306 like the other screen. Have to use a different library and everything. Guess it's not going to be as simple as just plugging in a bigger display.
New library needs you to define which controller it uses:
https://www.aliexpress.com/item//32844104782.html
from this list:
https://github.com/olikraus/u8g2/wiki/u8g2setupcpp

Doesn't say on the product page, any idea?

[Anonymous No.1473844]

>yeah I've built my own computer.
>nice, me too, it was kinda hard but worth it in the end
>You thought it was hard? Lol. Mine has a xXGTX1337 TITANXx with THREADFUCKR BARRACUDA

>the hardest parts were writing the bootloader and usb-serial communication driver and planning out the memory map. What CPU did you use? I chose a motorola 68000 for its 32 bit data bus and because it's easier to cross-compile for
>oh, you meant... but m-motorola doesn't make CPUs t-though they make phones...
every fucking time

[Anonymous No.1473847]

>>1473820
I think each semiconductor company had their own names for each signal and as they developed one or another interface they used their own distinct names (NXP: I2C which has SCL and SDA, Intel did SMBus, Motorola did SPI (I forget which but I know there you'll see SCK), and so on). If you're having problems, then you should always be checking the datasheet for the signal in question. Best of luck to you.

[Anonymous No.1473894]

>>1473750
write some code and count some cycles. you'll probably find the amount of time a single scan takes is independent of the data, or at worst within one or two CPU cycles per pin per turn. if you use a hw timer as your timebase and you wait on it between turns, you should have no real problems with synchronization

>>1473784
STM8s are 6502s with pretensions. not the best for asm but they might make an ok C target. btw you can also get STM32F0 in SSOPs for under a buck

>>1473791
you supposed wrong. you need only one MOSFET and a couple of pull ups per line, Pic related

>>1473844
>from the component level?
>yep
>ah, cool, what motherboard did you use?
mmmmyep

[Anonymous No.1473937]

I got a lot of these things and i want to put them into walls instead of regular light switches. (the description says it too)
https://www.aliexpress.com/item/TTP223B-1-channel-Jog-digital-touch-sensor-capacitive-touch-touch-switch-modules-Accessories-for-arduino/32616899749.html?spm=a2g0s.9042311.0.0.27424c4dFViKQ8
But i have a problem. I connected the sensor to 5V and measured the voltage on the signal pin.
When i touch the sensor the singal goes from 0V to 5V. This is correct. But the problem is i actually have to physically touch it, which is useless, since i want to have it INSIDE the wall (about 1 cm deep), since i obviously don't want to have a bunch of ugly-ass ICs glued to my walls, at that point i could just keep regular switches.

So how do i make this cunt, dettect human hand presence at 1cm?

[Anonymous No.1473949]

>>1473937
you don't. somehow you'll have to put the sensor plate closer to the surface (they work fine through a couple of mm of ABS). that's more of a mechanical design problem, depends greatly on your available tooling/materials and access to the location you want to place the switches

[Anonymous No.1473952]

>>1473949
>compromises
Disgusting.

I got a light bulb above me head, grabbed a wire, piece of aluminum tape and BOOM 1CM detection distance through a thick piece of wood no problem.

Never settle.
Persevere.
Follow your dreams and you too can turn on lights from 1cm away one day.

[Anonymous No.1473954]

>>1473937
>touch-sensor-capacitive-touch-touch
You again? We had that before. RTFD, remove C1, increase the sensor area (it's on the backside). Maybe you arrive at 3mm. Good luck.

[Anonymous No.1473989]

>>1473537
>add a resistor in series with a MOSFET, all in parallel with R3. select MOSFET for appropriate turn-on voltage according to the logic level you'd be using with it

Thanks Anon. I tried to sketch your idea. Did I get it right? Is it okay to use ordinary NPN transistors like this?

[Anonymous No.1473993]

>>1473989
looks bretty gud
>Is it okay to use ordinary NPN transistors like this?
in this case, with a relatively high Vfb of 1.2V and no need for especially precise current adjustment, it should be okay

[Anonymous No.1474010]

Hey boys can anyone give me a quick rundown on FPGAs? We're using them at uni as part of our CS course, and they seem kinda interesting so I thought I'd get a dev board similar to the one we're using (spartan 3 something 144 pin) but idk what to look for.

[Anonymous No.1474019]

>>1474010
the best guide to choosing an FPGA dev board is to ensure it has the desired I/O devices, peripherals, memories, and ports, and an FPGA with enough gates to fit your design. many boards will have LEDs, buttons, etc. which can occasionally be nice. many dev boards have connectors for wider parallel interfaces, some newer ones have PMOD (4 signal + power + GND) which is a popular connector interface for serial peripherals of many kinds
Spartan-3 is kinda old, so you might have best luck finding a board on ali or ebay, and you'll probably need a legacy version of the Xilinx software to build bitstreams for them
also, Xilinx software tools prefer their proprietary programming hardware. you can find USB Platform Cable clone programmers on ali/ebay for $30 or so. they're not necessarily blazing fast but you should be able to use them for debugging with the Xilinx tools
the Bogdanovs are rumored to have a universal FPGA programmer that works for all vendors and software past and future, but nobody has seen it and lived to report back

[Anonymous No.1474033]

>>1474010
qrd: FPGAs are small-to-huge gate arrays to implement any logic you want, including processors. But the real benefits are from being able to implement custom peripherals, or heavily parallel logic tailored to your specific application, or very high speed computations.

[Anonymous No.1474042]

>>1474033

Acronym Definition
------------ ------------
QRD Queer Resources Directory

[Anonymous No.1474051]

>>1474042
take the bogpill, or else

[Anonymous No.1474089]

New to RF and I want to explore it more as I can't take the courses that I would like to right now. What are some software I should get familiar with for RF and Microwaves?

[Anonymous No.1474093]

>>1474089
/ohm may not be ideal for this, though I'm not very sure. There was some guy who cleared my misunderstandings by using some magnetics software, but chances are EM waves are calculated in different software to static magnetic fields. Someone on /ham might know, or have a hunt through stackexchange and wiki answers.

[Anonymous No.1474099]

>>1474093
I don't believe there is a /hrg/ running rn

[Anonymous No.1474102]

>>1474099
That is correct, did it hit bump limit recently?

[Anonymous No.1474111]

>>1474102
aye
>>1444577

[Anonymous No.1474113]

>>1471129
>>1471130
>>1472918
eh, you could go on Udemy and get a quick course on embedded design, depending on the size of the company you're applying for. The larger the company, the less amount of knowledge you need since the business has a guy for each step. like a hardware guy/ a power electronics guy/ a low level electronics guy/ an embedded guy, maybe a DSP/matlab guy/IT guy. in small companies you'll need more knowledge where the degree is really going to be used. Like a lot of position I've seen, which I'm gunning for requires you to be the hardware/trace layer/power electronics/dsp/embedded which requires you to have done a lot of homwork/projects.

Electrical is a mixed bag, it's really about the location you wanna work. If you're outside a major city the majority of the jobs that exist are for construction so basically drawing wires from a 3rd party electrical panel to field devices in autocad, which is super autistic and very limiting. You'll be working for either the consulting company that says you need to have these outlets or these fire alarm devices in these locations to meet spec or the contract company that wins the bids for the consulting company.

There's very few embedded design jobs like the one i'm going for, but they exist. For you, I'd recommend going for either since they both overlap and just get the Principles books, Mims is top tier, and your professors would love the fuck out of your research reports if you adapted his style of representing designs. Work on projects constantly, women are a waste of time, and ask your professors about your ideas. A fun mini project you could do is go through the book: Hardware Hacker by Andrew 'bunnie' Huang or his other book Hacking the Xbox.

[Anonymous No.1474128]

>>1474089
What's are some prereq knowledge needed to read that?

[Anonymous No.1474160]

test

[Anonymous No.1474186]

test

[Anonymous No.1474260]

Are we back yet

[Anonymous No.1474274]

test

[Anonymous No.1474279]

Why is EEVBlog's dave jones such an insufferable sperg?
>watching mailbag video
>someone sends him an open source usb scope
>oh boy here we go
>it is as expected
>whiny voice
>full of contempt
>complete evisceration
>the most trivial of constructive criticism
>legion of fanboys lining up saying "i like that he calls a spade a spade"
I feel bad for the guy. He was just trying to make a thing and then here he is with someone he probably looks up to in some way, gets eviscerated completely, and a bunch of parroting fanboys shitting on him some more
It would have been sufficient to tell him this isn't an "oscilloscope" and some things he needs to change
But nah just eviscerate it like its some overpriced commercial product that don't do shit

[Anonymous No.1474285]

>>1474260
I think we're fine

[Anonymous No.1474369]

If i have a digital temperature probe that uses 1wire protocol (actually 3 wires) connected to arduino, what is the maximum length of the wire i can have and still get reliable readings?

[Anonymous No.1474382]

>>1474369
Have a look at the datasheet of the sensor. There should also be documentation of whatever protocol that is, and probably other arduino-users' similar questions. If you use coax with the centre conductor as the signal wire you'll probably get better range than otherwise, but it's probably not worth it.

[Anonymous No.1474384]

>>1474369
Guidelines for Reliable Long Line 1-Wire Networks

[Anonymous No.1474388]

How likely is this to explode?
You take apart a psp battery to get the charging chip, wire in your own lipo to it and remove the disk tray to fit it in.
I can't see anything wrong with it, especially if you use a larger capacity battery that would take higher charging current.

[Anonymous No.1474389]

Why is it so hard to make a crystal oscillator oscillate on a breadboard. I built a classic Pierce oscillator with an 8MHz crystal using at first a 3.3M and later a 1M load resistor. I tried a load capacitence of 22pF and 33pF and even whacked in some reverse biased 1N4148's with about 1pF of junction capacitence instead of capacitors and I just couldn't get it to start. I tried it with a 1k resistor in the position of R2 in the pic and I tried it without it and still couldn't get it to start. I kept all the lead lengths as short as possible. I tried using 74LS04 for the inverter, then I tried a 74LS14 instead. No matter what I do I just cannot get it to work and I don't know why.

[Anonymous No.1474394]

>>1474389
Often a variable capacitor is included in the circuit diagrams I see, perhaps throw one of those in there.

[Anonymous No.1474428]

>>1474389
A Schmitt trigger inverter may not provide the linear loop gain required for start up. Better use a normal inverter like a 74HC04 or similar. Pic: Unbuffered 4069 (74HCU04 version if available) for an adjustable 32kHz clock oscillator and the equivalent transistor version. Note the large series resistor typical for low frequency crystals to limit the load.

[Anonymous No.1474497]

>>1474388
3.5% likely. you should sense the battery temperature to prevent overheating during charging

>>1474389
>on a breadboard
you mean, a solderless breadboard, with loads of capacitance all over the damn place?
smug_anime_girl45616.jpg
I recommend you go directly to breadboard

[Anonymous No.1474518]

my phone headphones' right speaker stopped working(or so i thought)
turns out it still werks but is just 98% quieter, also the mic doesnt work
probing further i discovered mic to fourth pin gives normal sound but ground to fourth does it at 2% volume
im guessing the wires got fucked up somehow

what would be the easiest way to find where it got fuked?
or make the phone use mic to fourth for the second speaker

[Anonymous No.1474522]

>>1474279
>connectors so close it's borderline unusable
>horrible mess of a schematic
>software written in pyshit 2.7
That scope is the result of a successful crowdfunding campaign, so he was shitting on a commercial product and not just some guy's little pet project. Giving a product a free pass just because it is open source is the exact opposite of what you should do if you care about the open source community and quality open source hardware.

[Anonymous No.1474541]

>>1474522
sure, they should have gotten some design reviews before manufacture, but give 'em some credit for getting a thing through the product process, at least
also python3 adds what, exactly?

>>1474518
not worth fixing. buy a new one

[Anonymous No.1474572]

>>1473993
I assembled the circuit on protoboard (without Q2/Q3) and it indeed works.
With the component values shown on the image (I only have a 22k trimmer) a 8.2 ohm shunt resistor seemed to produce the desired output current of 250 mA.

However I found that the dropping resistor wastes a lot of power. With 250 mA of output current, there's over 2 V across the shunt and 0,25 A * 2 V=500 mW of energy wasted as heat. I think this is a kind of disappointing result.

Would it be possible to use a transistor amplifier circuit to amplify the voltage developed by the shunt resistor? That way a lower resistance shunt could be used, decreasing its power consumption.

[Anonymous No.1474581]

>>1474541
>also python3 adds what, exactly?
Well, it isn't on its death bed for one thing.

[Anonymous No.1474604]

>>1474497
>you mean, a solderless breadboard, with loads of capacitance all over the damn place?

That shouldn't really matter that much, I've seen other people use crystal oscillators on breadboards. Maybe above 10MHz yeah okay that'd be more of a problem but I think it's possible to get away with 8MHz. I was using a 4-pin 7.3ish MHz oscillator with all the oscillator circuitry in the can initially but I think I blew up the crystal when I put a large cap that was charged in the wrong holes and I think I caused a voltage transient that blew a hole straight through the crystal. Anyway before I killed it that crystal worked fine, I can't imagine going from that to the 2 pin 8MHz crystal with all the oscillator circuitry external to the crystal being a deal breaker.

Also my understanding is the load capacitence isn't even that critical as far as making the crystal oscillate is concerned. In parallel resonant mode as long as you have enough to provide the full 360 degrees of phase shift the value isn't important and only will determine the exact frequency of oscillation. I don't care if it's dead on 8MHz or a bit off so it's not important to my application. In series resonant mode my understanding is you don't need any external capactience at all.

[Anonymous No.1474636]

>>1474572
What's the point of regulating the current rather than the voltage?

[Anonymous No.1474640]

>>1474572
consider eliminating the trimmer and connecting the base resistor of Q1 to the other side of Rshunt. thus you only have 0.65V across it and 0.65 * 0.25 = 1/6W lost in it, which is pretty good
>transistor amp
I can't think of a clean easy way to do it. I'd probably try an op amp. since common op amps don't work very well near the top rail, I'd move the sensing to the low side

>>1474581
it still runs, I fail to see what the problem is
>not on its death bed
languages don't need to add complexity aka features for their own sake to remain useful
did you take a wrong turn on the way to >>>/g/ ?

>>1474604
>using crystal oscillators
there is a huge difference between using a module and building a circuit
the inverter's input and output pins are probably on adjacent rows, correct? I just measured a solderless breadboard I had around and got 1.8pF between adjacent rows. so 1/(2*pi*8MHz*1.8pF) = 11kohm of negative feedback at frequency
also, don't use a TTL inverter for this as the input impedance is far too low. you're probably swamping your crystal's signal

[Anonymous No.1474663]

>>1474640
Wait will a 74LS14 not work on a proper PCB either? Not him, just have another project.

[Anonymous No.1474668]

>>1474663
I'd prefer to use something with a higher input impedance. 74HC is about right for non-overtone crystals. there are unbuffered inverters designed especially for oscillating crystals

[Anonymous No.1474670]

>>1474668
Yeah but I already bought the LSs, plus the rest of the PCB is using LS series. Can I just use 2 MOSFETs instead?

[Anonymous No.1474674]

>>1474668
Original anon here, can I use an op-amp in an open loop configuration or a comparator instead? I also only have LS series TTL and I don't think I have CMOS inverters either.

[Anonymous No.1474675]

>>1474674
If the opamp or comparator can handle the frequency then probably. You could even make a sinusoidal oscillator with a closed-loop op-amp if you know what you're doing.

[Anonymous No.1474677]

>>1474674
most common op amps would be too slow (gain-bandwidth product somewhat less than 8MHz)

>>1474670
>>1474674
both of you, go for the transistor implementation here
>>1474428
the 2N3904 has plenty of gain-bandwidth for this (300MHz). its output will probably be strong enough to move the LS14 back and forth

[Anonymous No.1474678]

>>1474675
Ah shit I forgot about that. TL072 unity gain bandwidth is like 2MHz I think, no good. It'll have to be a discrete transistor solution I guess then.

[Anonymous No.1474679]

>>1474640
>it still runs
"Hurr durr who cares if it's written in cobol/c89/brainfuck as long as it still runs, am I right?"
>I fail to see what the problem is
Yeah, why write the OPEN SOURCE software for your OPEN SOURCE hardware in the more popular, more supported and newer version of a programming language when you can just use the old one for no fucking reason? I mean it's not like people being able to modify the code is one of the selling points or something.
>languages don't need to add complexity aka features for their own sake to remain useful
Well idk about python but most languages add features that, for the most part, make the resulting programs less complex, and I'm certain python is no exception. You'd have to be some kind of giga-brainlet to think it's a good idea to sacrifice code simplicity for language simplicity, or to worry about the bloatedness of python 2 vs 3 (it's like ebola vs aids).
>did you take a wrong turn on the way to >>>/g/ ?
Preferring python 3 over 2 is not some obscure opinion confined to /g/.

[Anonymous No.1474681]

>>1474522
Nobody said he should be given a free pass
Where are you even getting that from
he went beyond criticism and all but said he should stop trying. Criticism wouldn't be so full of contempt
if its just crowdfunded its not exactly commercial now is it

[Anonymous No.1474683]

>>1474681
>Nobody said he should be given a free pass
The poster I replied to was basically saying that he should have gone lighter than usual on the guy simply because he's a hobbyist, whether that counts as a "free pass" is just semantics.
>he went beyond criticism and all but said he should stop trying. Criticism wouldn't be so full of contempt
Well it's his usual style when it comes to reviewing less than adequate products, and as I said, I think it's his job to hold all products to the same standard regardless of who made them.
>if its just crowdfunded its not exactly commercial now is it
Idk how some random nigga ripping you off on kickstarter is better than chinks ripping you off on ebay or whatever.

[Anonymous No.1474713]

I managed to measure some parameters of my 8MHz crystal. Is there some kind of formula to plug this data into in order to find the values of external components needed to make it oscillate properly? My test fixture's a bit crude and I only did a thru calibration instead of a 3-term so these results might not be dead on but it should be close enough.

Equivalent CKT 4 is on screen and the resonant 0 phase measurement gives:
Zr = 670.5MΩ
fr = 7.998558220MHz
Za = 212.5 µΩ
fa = 8.006167269MHz

[Anonymous No.1474741]

>>1473952
a rare breed: the /diy/er who actually goes and builds something instead of arguing about building things.

[Anonymous No.1474743]

>>1474388
do you really have to use that original PSP charging chip??
there's only 3 wires coming out of it. is it really trying to do any kind of communication with it?
before you do that, make sure it's really necessary.
you can get little TP4056-based lithium-ion charger modules that are cheaper and more reliable/controllable than that hackjob.

[Anonymous No.1474745]

>>1474683
>The poster I replied to was basically saying that he should have gone lighter than usual on the guy simply because he's a hobbyist, whether that counts as a "free pass" is just semantics.
So not being a vitriolic cunt is "giving a free pass" now? Okay
>Well it's his usual style when it comes to reviewing less than adequate products, and as I said, I think it's his job to hold all products to the same standard regardless of who made them.
His usual style sucks, and its worse because he has an echo chamber of fanboys who hang on his words like gospel
>Idk how some random nigga ripping you off on kickstarter is better than chinks ripping you off on ebay or whatever.
Implying that's what I said
It's a one man show without some massive engineering team behind it. You're the only one here saying its to do with money

[Anonymous No.1474746]

>>1474741
Not him but I'd build shit if i had money to buy a ton of extra components to tinker with and learn

[Anonymous No.1474747]

>>1474743
>before you do that, make sure it's really necessary.
>Sony
it would be out of character for them to not check for it

[Anonymous No.1474748]

>>1474745
agreed. his videos really should be about 25-40% as long as they are

[Anonymous No.1474782]

I'm trying to find any article for lithium ion batteries, particularly storage ones, where you overbuild them for longevity. I tried searching overbuild, overcapacity, longevity, some other bits, but there wasn't even an article CONTAINING the concept.

An example being I read Tesal's Powerwall 2 had a nominal capacity of 13.5 kWh, but the battery was actually 20kW. Simple reasons: batteries wear out faster at higher temps and time. Higher-than-listed capacity means less resistance for a given capacity and power draw, a decrease in capacity is adjusted for in software over time, etc. I don't remember hearing they used capacity modulation, to keep the max charge within a more efficient window (how batteries are supposed to last longest around 70-80% charge, not 100%).

The ultimate goal was instead of battery lasting like 3-5 years, they could guarantee it'd last 7 minimum or whatever, *at the nominal power and ratings*.

[Anonymous No.1474811]

>>1474782
>I don't remember hearing they used capacity modulation
I do remember hearing something like they locked out the top and bottom 30% on the car battery packs. this seems sort of relevant http://www.freepatentsonline.com/8629657.html
in general, things Li+ don't like, most disliked first: physical damage, heat, cold, overcharging, undercharging, charging while cold, overdischarging, charging, discharging, being looked at funny

[Anonymous No.1474818]

>>1474782
Perhaps look into the differences between high-C low-capacity (like vape cells) and low-C high capacity. LiFePO cells and those whacky ceramic cells are also worth looking into. Did you look about on Scopus?

[Anonymous No.1474820]

>>1474636
I found that constant current drive produces a more gradual turn-on of the lamp.

>>1474640
Thanks again I will try the low-side current sensing next. I was already doing some research about high-side current sensing with an op-amp, but it seems to require a specialized IC to work well.

[Anonymous No.1474829]

>>1474384
How is a squiggly line supposed to answer my question?
No matter, I googled and found that maximum reliable length is 20m which is plenty enough for me

[Anonymous No.1474842]

>>1474743
>do you really have to use that original PSP charging chip??
No, but it's convenient. This way you can use the psp charging port and don't have to take the battery out or cut any extra holes in it to charge.

I just got the chip out of one of the batteries, the board had B+ and B- printed on it, but when charging with the cable 160mv comes out in reverse. Tried connecting B+ and - to my benchtop power supply at 3.6v and it wouldn't power on. Maybe it's looking for some characteristic of a battery like internal resistance or maybe the cell wasn't 3.6v and it's doing some voltage conversion on the chip.

[Anonymous No.1474844]

Anyone here make a diy soldering station that can employ JBC irons and tips?

[Anonymous No.1474853]

>>1474844
Is there a datasheet for those tips?

[Anonymous No.1474865]

>>1474853
Not that I know of
The two videos I saw just had them taking an ohmmeter to it and then guessing at the internal connection (they're probably right though)
but as far as characteristics of the heating element and of the thermistor, that they are definitely guessing at

[Anonymous No.1474868]

>>1474865
If you have a thermocouple lying around (and a multimeter you can use separately from it) then you can log the element resistance as a function of temperature. I think you'd also need a power supply that can provide power to the element at a variety of levels.

[Anonymous No.1474879]

>>1474820
>gradual turn-on
Did you consider a soft-start capacitor across R2? With R2/R1=39K/10K you get about 6V and you can dim the light by injecting a small current I.

[Anonymous No.1474885]

>>1474746
this is no excuse
hardware from china literally costs pennies and with free shipping too
all you have to do i literally not buy a cup of coffee one day and you have enough money for an arduino

[Anonymous No.1474891]

>>1474885
>chinese hardware
>arduino
No thanks faggot
I'll wait for my shit
I've already breadboarded some shit and i just want to learn other shit anyway

[Anonymous No.1474894]

>>1474893>>1474893>>147>>1474893>>1474893>>>1474893>14748934893>>1474893>>1474>>1474893893>>1474893>>1474893
>>1474893
>>1474893
>>1474893
new

[Anonymous No.1474898]

>>1474894
Thank you based anon

[Anonymous No.1475746]

>>1473952
Are those Chino-Junk Optical Thermometers actually any good?

[Anonymous No.1475782]

>>1475746
They show temperature so i would say yes and cost like 5 times less that from the retail jew