Thread cracked: >>2989599>I'm new to electronics. Where to get started?It is an art/science of applying principles to requirements.Find problem, learn principles, design and verify solution, build, test, post results, repeat.Read the datasheet.>OP source:https://github.com/74HC14/ohmOPbake at page 10, post in old thread>Comprehensive list of electronics resources:https://github.com/kitspace/awesome-electronics>Project ideas:https://hackaday.iohttps://instructables.com/tag/type-id/category-technology/https://adafruit.comhttps://makezine.com/category/electronics/>Books:https://libgen.is/>Principles (by increasing skill level):Mims III, Getting Started in ElectronicsGeier, How to Diagnose & Fix Everything ElectronicKybett & Boysen, All New Electronics Self-Teaching GuideScherz & Monk, Practical Electronics for InventorsHorowitz and Hill, The Art of Electronics>Recommended software tools:KiCAD 6+CircuitmakerLogisim Evolution>Recommended Components/equipment:OctopartLCSCeBay/AliExpress sellers, for component assortments/sample kits (caveat emptor)Local independent electronics distributorsladyada.net/library/procure/hobbyist.html>Most relevant YouTube channels:EEVblogW2AEWMoritz Klein>microcontroller specific problems?>>>/diy/mcg>I have junk, what do?Shitcan it>consumer product support or PC building?>>>/g/>household/premises wiring?More rules-driven than engineering, try /qtddtot/ or sparky general first>antigravity and/or overunity?Go away
>I have junk, what do?Use it to practice soldering and desoldering.
>>2996880Eyeballing the dimensions and using the density of barium titanate (which is apparently what ceramic caps are made of)...>(40cm * 20cm * 20cm) * 6.02g/cm**3 = 96.32 kgThat picture is total bullshit, there's no way she could lift that, especially not with that grip
>>2996936If you look closely you will see that they're schoolgirls practicing board assembly using painted foam blocks and Velcro to stick them on the PCB. Also, it's a drawing and nothing is real, Percy.
been fiddling with a Wien bridge oscillator today, trying to get the amount of distortion down. -80 dB THD here, a huge improvement of my previous attempt at -40 dB or soevery now and then, modulo breadboard connection crappiness, that second harmonic is down in the noise floor, so -100 to -120 dB. the trick is to bias the JFET just right so that it only has to servo the gain in a narrow gainthere's circuits that can go as low as -140 dB but I can only use parts that are roughly 0402 size in my application. this means no LDRs
>>2996943forgot the circuit
>>2996951Put another diode in anti-parallel to D1.
>>2996953no, that'll ruin the AGC function. D1 acts to pull the gate of Q1 low when the oscillator reaches a particular amplitudean earlier version of the circuit did have an antiparallel diode, but with a 1M series resistor. this was to pull the gate voltage up slowly. putting a resistor to the source of Q1 proved simpelr
>>2996936No it's not a massive PCB, rather the girls are tiny, and maybe they're even robots. As per the square cube law, they'd be extremely strong for their size, much like insects. With one on each corner, they could likely lift that QFN, but it would be more fun to imagine them wearing hard hats and directing a crane to move it into place.>>2996951Putting a resistor in series with D1 would prevent the gain changing as quickly, and would probably reduce distortion. But conversely it would reduce your amplitude control. I've never made a sine oscillator though so I'm just spitballing.I know some people make sine oscillators using an incandescent light bulb as the variable resistance, perhaps you could use a tiny thermistor? Like 0402 or 0201, so only a few milliwatts are required for it to regulate itself.Is it for a certain purpose?
>>2996958>Putting a resistor in series with D1 would prevent the gain changing as quickly, and would probably reduce distortionnope, tried that. makes things worse>I know some people make sine oscillators using an incandescent light bulb as the variable resistance, perhaps you could use a tiny thermistor?can't really do that, because I need to keep the power dissipation low. self-heating is a concern>Is it for a certain purpose?it needs to fit in a 3 mm tube (2.5 mm ID) because reasons. power and signal goes through an RG-178 coax. it's for basedence!
>>2996960Commit the circuit to protoboard just to get a baseline. I don't trust breadboards.
>>2996961yeah I just had the same thought. tomorrow
Why am I having such a difficult time finding stereo FM transmitter IC chips on Mouser? Something like the BA1404 or NJM2035. I want to take two audio channels and transmit them over FM at low power. Just like BT FM car radio transmitters.I've looked in the modulators section, the RF transmitter section, and the audio transmitter section. Do I have to have a computer generate the signal instead?I do not have a great understanding of modulation and wireless transmission. Ideally I wouldn't need to in order to use an IC chip for the job.>inb4 FCC needs license and registration for thatLet me worry about that
Volume box done. Input jack, output jack, and a headphone bypass on a switched jack.
>>2996983>Do I have to have a computer generate the signal instead?before bluetooth, there were hundreds of FM transmitter models being soldfrom super-cheap ones with fixed frequencies, to sophisticated ones with digital frequency display, and MP3 decodingi got half-a-dozen from the thift-store for around $2 eachfind 'em on ebay, tooit's ridiculous to make your own; you''ll never match the price and features
>>2997000Still worth designing if he wants anything other than the default antenna or input filtration. Maybe he wants to add an amplifier or BNC or SMA socket, or maybe he wants to ditch the high-pass input caps and transmit a DC or very low frequency audio waveform. Or maybe he just wants to optimise the design to something with a lower noise floor.I’ve got the PCB from a cheap product (inline headphone remote) inside one of my projects, but it frequently malfunctions in ways I do not understand. Even just knowing how these things are designed and what their limitations are can be very useful.On that note, if anyone can figure out what chip is inside these apple-style 3.5mm earbuds that would be appreciated. I couldn’t find anything, so now I’ll try looking into the other end, the chips behind the TRRS sockets in cell phones, to see if they describe the protocol. Maybe there’s info in the Pinephone or other open source device documentations, or maybe there’s a relevant chip name in an ifixit teardown.
I’m not exactly an expert but I can do some arduino stuff. I gutted a broken robo vacuum and want to re-use some parts. I identified and tested most but wondered if you could help identify the others and say if/how theyre usableFirst one is the suction motor. The only motor that doesn’t start to spin when I put 12V on it, tried both polarities. It has 2 large and 2 smaller wires (I think Hall effect sensor). Does this need PWM to run? No markings visible without breaking the fan off, except ‘Nidec’
>>2997034Then there is this one. Sensor seems very simple but it’s the only sensor that comes with its own complex PCB. It used to look down at the floor, it says HTW225A but that doesn’t yield any result. What could it be?
>>2997035Finally a camera with a ribbon cable. The camera itself in the housing is a small cube shaped module with zero markings. Was wondering what to connect this ribbon to, or what kind of protocol this would be using
>>2997034I'd guess that there's a motor driver on the board, and that that's a 3 phase brushless motor. The driver would be activated by talking to it via the comms wires, maybe it's I2C or SPI or UART, or just PWM or even an analog voltage. Maybe it's like a 4-wire PC fan, and one wire is tacho and the other is PWM, maybe it just needs 3.3V or 5V or whatever on the PWM pin to turn on. But I'd hold off on putting voltage on those small signal wires especially, very easy to damage stuff.By feeding it negative voltage you probably fried the driver, but it's possible you didn't. To do anything with it, you'll want to reverse engineer that PCB, should be very feasible if you can actually get it out of that housing.>>2997035Where was the sensor? I'd guess maybe it's an ultrasonic transducer, for distance-measurement in addition to the camera(s).>>2997036That one you're either going to have to reverse engineer the camera by tracing wires back to the main ICs inside the housings, or by tracing the other side of where that ribbon cable connected to on the vacuum's main board. There looks to be 4 differential pairs, it's probably something like LVDS.Ideally you'd want to use an oscilloscope or logic analyser to see what happens on the wires as the full vacuum operates, or at least attempts to operate.
>>2997037Thanks a lot anon. I’ll try 24 and 5V or 3.3V PWM on the small wired. It seems a bit like the motor holds itself in place when applying just 24V. I’d have to break it to see the board and it may be broken already so if that doesn’t work I guess I’ll scrap it> Where was the sensor?Bottom of the machine facing down at the floor. I’m mostly curious as to what the PCB does and why this is the only one they didnt connect directly to the main board The camera ribbon is just a tiny bit too large for the rPi camera port (2 pins too many I think) but looking at the ribbon it looks like only 6 are connected (the rest looks like either ground or isolated) but maybe in missing something
>>2997000I want to integrate it into a portable music player. I could just buy some old BA1404 on eBay, and that might be the best solution. I was just hoping there is something equivalent available through Mouser
>>2997041>I’d have to break it to see the boardOk then at least do some measurements with your meter, to see if there's diodes between the input power pins (e.g. the body diodes present in an H-bridge or 3-phase inverter). And use current limiting resistors on your PWM signals (1-10k should suffice).>facing down at the floorNot sure what it would be. Maybe something to do with the charging dock? Or a sensor to discriminate between flooring types?>looking at the ribbon it looks like only 6 are connectedFrom bottom to top:>1 - GND>2+3 - differential pair>4 - GND>5+6 - differential pair>7 - GND>8+9 - vias going to other side, possibly another differential pair>10 - GND>11 - via going to other side, probably a control signal>12+13 - GNDs>14 - via going to other side, probably a control signal>15 - GND>16 - NC>17 - via going to other side, probably a control signal>18 - GND>19+20 - vias going to other side, possibly another differential pair>21 - GND>22+23+24 - NC>>2997044Gonna make it the right size to fit in a cassette deck too, with magnetic transducer pickups along the top? It's only a matter of time before someone makes an extra thin MP3 player that beams data into a CD player's optical pickup.
>>2996998And here it is flipped and mounted under my desk. Next to my audio mixer.
>>2996983what kind of range do you need from your transmitter? what kind of CPU do you have in your device? if you have a PWM channel that goes up to say 100 MHz then generating FM is quite easy. if it doesn't go that high then you can use a lower frequency like 33 MHz and pass the signal through a diode, then through a band pass filter to extract the 3rd harmonic
>>2997048Broadcast FM varies the frequency by up to +/-75kHz modulation depth. For 16 bit audio, a least significant bit would be 2Hz of modulation. To do that digitally would be extremely difficult, requiring that your PWM channel be clocked 2^16 times faster than the output 100MHz.An MCU with a 400MHz clock can produce 400, 200, 133, 100, 80, 67, etc. MHz from its internal timer/counter, it’s infeasible to generate pre-modulated FM.Instead you want a PLL. Digitally, you could hop around multiplication and division registers to create arbitrarily finely divided frequencies, but you’ll probably still get glitches. For FM, a conventional VCO is probably the best way to go. For stereo, that requires analogue circuitry to amplitude modulate the difference of the channels on a supersonic carrier, and add that to the sum of the channels, before piping that into your VCO. Seems like a fun project, but a big time sink.
>>2997050https://github.com/cnlohr/channel3https://github.com/cnlohr/avr_vhf
>>2997050>what is phase modulation
>>2997053CNLohr’s videos on generating NTSC from the ATtiny and 8266 just show him generating black and white video, which is fixed frequency.But his video on generating LoRa chirps with an ESP shows that semi-continuous frequency sweeps are possible, by generating varied lower frequencies and relying on their higher square harmonics. You could get dozens, maybe hundreds of frequencies from this technique, more than enough for a proof of concept of FM broadcast radio, could probably even make it stereo and/or dither the frequency to give a higher effective number of bits. But I doubt you’d approach 16-bit audio quality.>>2997054I’m not sure I understand how phase modulation or phase shift keying could be relevant to outputting broadcast FM from a microcontroller. Might you enlighten me?
>>2997117you can twiddle the PWM phase using the derivative of the audio signal. push the phase forward when the derivative is positive and vice versa. you can add dithering to get more bits. this probably consumes most if not all cycles in anon's CPU, unless it has more than one core
>>2997118If your PWM counter is going up 100 steps in a triangle wave, and the counter is clocked from the main clock directly, then that’s 200 different phases, which isn’t very much. It also requires that your system clock is 100 times faster than your PWM output, so it would still be infeasible to directly output that 100MHz broadcast FM signal. With downsampling techniques it’s possible.The limitations would be different to changing the output frequency conventionally, you’d get the full dynamic range up at the 15kHz range, and down at 150Hz you’d have 100 times less range. And because it’s still outputting edges synchronously with the main system clock, you still have the same fundamental resolution limit as just adjusting the counter’s output frequency. Maybe they can be dithered differently, and maybe it’s easier to subsample, but I doubt it.Some MCUs like the ESP32 have a built-in PLL piped to output pins that would be better for this than subsampling alone. You might instead be able to mess around with clock calibration registers to output a continuously varying square wave.
>>2997129perhaps generating a baseband signal and then writing a tightly optimized mixer loop that said signal is fed into could work? that is, one could use a GPIO pin as a 1-bit ADC and push SDR samples to it. such a scheme could be expanded to say 8 bits with an R2R ladderit sure would be good if anon told us what he's working with, CPU-wise
need a sanity check please, is that op amp upside down to buffer the square wave coming in at pin 9?
>>2997158>is that op amp upside downyepreverse + and - inputs
>>2997161it works either way up in Falstad, but is this one of the cases where simulation doesn't reflect reality?
Yesterday I sold my old soldering station and today I sold my hot air station. And my new ones haven't arrived yet and won't arrive for another week at least. I feel cortisol spiking, /ohm/.
>>2997158This original configuration makes positive feedback, producing a comparator with hysteresis. Any input wave with sufficient amplitude will produce a square wave output, while a low amplitude input wave will produce a DC output. In reality, if that’s a single-supply rail circuit (e.g. 0V GND to Vee and +12V to Vcc), it probably won’t work as a proper comparator. If this is intended, it’s somewhat odd to be using an op-amp instead of a dedicated comparator IC.If you flip the op-amp, it will instead act like a non-inverting amplifier with a gain of 11, and will work well regardless of the supply rails. Though if the input wave’s voltage ever goes above Vcc/11 or so, it will see clipping distortion on the amplifier’s output.Considering the input is something akin to “square raw” and the output is “Clock 1”, a digital output of a comparator circuit seems more sensible than an analog output of an amplifier. So I’d suggest the original orientation is intentional and correct.For confirmation you’d be trying to answer the questions “what task is this amplifier circuit supposed to be doing” and “are the power rails and input signals correct to facilitate this task in one of the circuit orientations”.>>2997163Use spice, or some other simulating software where op-amps have supply rails, input and output voltage ranges, and can clip.>>2997168How are you going to fix your new soldering station when it dies if you don’t have a backup?
>>2997170>How are you going to fix your new soldering station when it dies if you don’t have a backup?After I listed them for sale, I was expecting that it would take me weeks to sell. Instead, both were gone in 2 days. I recovered about 70% of what I paid for the new ones and I'm happy about that so all the hassle of not having any tools is worth it. I'm just an amateur who does it as a hobby. Anyway, poor people problems.
>>2997173Mind telling us what you sold and what you bought?
>>2997174>Mind telling us what you sold and what you bought?I sold a Weller soldering station and a Yihua hot air station. I bought replacements from China. Got a battery-powered soldering iron that uses C210 cartridges and bought an Atten hot air station.I didn't spend a lot of money and feel like I'm upgrading by a lot.
>>2997176>battery-powered soldering iron that uses C210 cartridgesHow many watts? I haven't seen a battery iron above 30W, and most of the compact ones are down at 8-15W.
>>2997199>How many watts? I haven't seen a battery iron above 30W, and most of the compact ones are down at 8-15W.28W on battery and 65W when plugged in over USB-C. Given the specs for C210 cartridges, it has enough juice to drive them well. Everything I use an iron for, this can do it. After looking at various cartridges, I went with C210 since they're like 2ohms and heat up insanely fast.Last month I was at a local makerspace to get a piece of acrylic laser cut and while there I tried few of their tools. They had few battery irons and I've been lusting over one since then. Having one less wire on the desk, especially for something I use all the time, is worth the loss of power (most of which I don't even need).But your mileage and use case might be totally different of course.
>>2997225Link? I can’t find any rechargeable irons that come close to those specs. Better not be a fanttik t1, saw a review where it was like 80C too hot.
>>2997258>Better not be a fanttik t1,kek, no. it's a (better?) version of JBC's B.IRON 100.https://en.alientek.com/Product_Details/80.htmlI bought it the moment I saw the specs. I haven't seen many reviews of it (most were in Russian I think).
Before I burn my house down, I got a 24v 600w psu who's power switch seems to have failed closed. When I push the switch to turn off power it almost springs back to the on position and even if I get it to stay in the off position its still not off and power is being supplied. I just opened this out of the box and have only hit the switch like 5 times. Is this an issue with the psu or am i being an idiot somehow. Secondly Im trying to power like 5 or 6 dc motors possibly multiple at a time but its not worth a house fire and I can get away with running two at a time. I just need to know how much power they are drawing so I can get a psu with a current limits thats just 1.25 that. I thought about just using a multimeter but the concern is that the motor controller im using might try to brake the motors and use significantly more power during that time. Should I get watt power meter plug or is there something more inline I can use besides say an oscilloscope .
I think I worked out why my board (once again) doesn't work - it's not going to go very well sending +12 V signals into a chip powered off +5 volts, is it.This is a stupid error from when I was translating my breadboard prototype into a schematic, but it's got me wondering, how do you guys do it? How do you make a schematic and order a board that works flawlessly the first time around, with no mistakes?The best plan I have going forward is to build the prototype, finalize the schematic, and then re-build another prototype from scratch exactly following the schematic that I've made, that would seem to be a good way to catch errors.Some days I think I should have just stayed as a Python coding babby rather than trying to wrangle actual electrons
>>2997261Damn that looks nice. Might buy one for my brother, he doesn’t really like his current iron.>>2997280It’s not a software power button? I’d disconnect the thing from mains and measure continuity through the switch. And replace the switch anyway because switches are cheap.Are they brushed DC motors? The most they’ll ever draw is their stall current, which you can trivially calculate by measuring their DC resistance. If it’s below a few ohms, you’ll want to use a test current of 0.1A or even 1A. Or power it up with a low voltage, clamp the output shaft so it stalls and measure the stall current there, and calculate the DC resistance from that. The same sorta applies to BLDCs.But often these stall currents are stupid high and you never want to pump that many amps into a motor in the first place, at least not for more than a second. So instead you should be looking at a motor driver that measures current and prevents stall and short-circuit conditions from frying the motor. If you want to use a simpler motor driving scheme like simple MCU-controlled PWM, then I’d just look at the motor datasheet, there should be a continuous current rating. If you can’t find a datasheet for it, look for cheapish brand name motors of equal dimensions, voltages, and speeds.>>2997298Yeah analog switches are strange, you should familiarise yourself with the internal transistor diagrams of parts you don’t understand.I design a schematic first, it helps me organise my thoughts. Anything I’m unsure of I’ll simulate or breadboard, but just little segments, not the entire circuit. If it’s just something like not knowing a resistor value, I’ll leave testing that to the final build. Be that strip-board or a fab-house PCB. When getting boards manufactured, I’ll leave uncertain parts as THT, with jumpers and spare footprints as needed. Even if I’m getting SMD assembly. Often I’m not forward thinking enough and have to do bodges or board revisions.
>>2997298>how do you guys do it?we were all noobs once, sticking toy motors into 120V socketsbut we learned from our mistakesso, after 20 years, we are now godlike: making 10% fewer mistakes than when we were noobs
>>2997316>Might buy one for my brotherWait for my review. If it's any good, I'll let you guys know. I still think that getting that SI012 Pro Max is a much better deal and a much more capable iron (it can def solder XT60 connectors which this battery one will probably fail at). More versatile too. If you don't mind a wire, it's a much better buy too. With that soft liquid silicone 240W USB-C cable, it's an unbeatable setup.
>>2997320Why the Pro Max? Isn't it just a Pro with RGBs? Or is it capable of 100W while the Pro can only do 65W? I have a Pro, but only 8 ohm T12 tips, not the lower resistance shorter TS-style tips (second from left) that Sequre call "SI", but I've also seen called "ST" by others. I wonder if an Si012 Pro will output 100W if I get a lower-resistance tip into it?This picture also suggests it works with the SH72-style of shorter T12 tip (or T65-style), but they have a collet flange further up than a TS and further down than a T12, so I can't see how they'd seat properly. They only have two sets of electrical contacts from what I can see, so I think you could only use it with these SH72-style if you locked it in place with the M2 screw. Doesn't seem ideal, I'd like it if they made an iron with three sets of contacts. While the TS and SH72 contacts would probably be too close together, you could make their flange sit at different depths since the TS tip flanges are much wider.Anyhow, my brother specifically wants a battery iron (he already has a T12 station that runs on an M18 battery), and his birthday is in a month.
>>2997336>Or is it capable of 100W while the Pro can only do 65W?Correct. Pro Max can do 100W and Pro caps at ~65W. A different tip won't help you reach 100W with Pro.That battery iron will never be as versatile as wired irons. It's OK only for light hobby use with SMD and small THT components. They're definitely not good for an 8hr job type work since they last less than 40min on a full charge and take 9hrs to recharge. If you do some back of the napkin calculations, you can theoretically remove 8-9 large caps before it';s empty. Yeah, you can use it while plugged in but then you have a shitty wired iron.Get him a SI012 Pro Max and he will be much happier if he's a semi-pro.
>>2997339>A different tip won't help you reach 100W with ProConsidering it's 25V maximum, you're not getting above 78W with the DC jack, or 50W with USB PD, with 8 ohm T12 or TS tips. Only with the "SI" tips can it draw enough power.The Pinecil website describes TS tips as having 8 ohms, and the shorter ST tips as having 6.2-6.5 ohms. 25V @ 6.2 ohms is 100W, so I'm guessing the SI tips from Sequre are the same. Sequre's website describes the SI tips as "max power 100W", with no such labelling for its TS and T12 tips.That also means that with 20V PD on the Si012 Pro Max you're getting no more than 65W. I'm wondering what would happen if you put an ST/SI tip into a non-Max Si012 Pro. Would it PWM itself to limit power to 65W? Would it assume it's drawing less power than it really is, and force a 65W supply into hiccup protection? Would it detect a faulty tip and not heat up at all? Would it draw too much current and damage itself?Hardware wise, what could the difference be? Different PD negotiation chip or beefier power MOSFET, that's all I can imagine. And there might be no difference in hardware at all. There probably is software difference though.
>>2997351>'m wondering what would happen if you put an ST/SI tip into a non-Max Si012 Pro. Would it PWM itself to limit power to 65W? Would it assume it's drawing less power than it really is, and force a 65W supply into hiccup protection? Would it detect a faulty tip and not heat up at all? Would it draw too much current and damage itself?It PWM-limits itself based on calibrated resistance, not a 65W hard wall limit.I don't know about USB-C PD but I'm pretty sure that if you connect it to a 4S or 6S LiPo it would discharge massive amounts of current without caring about a 65W limit... and probably fry the P-MOSFET.>Hardware wise, what could the difference be? Different PD negotiation chip or beefier power MOSFET, that's all I can imagine. And there might be no difference in hardware at all. There probably is software difference though.An updated PD controller and software of course.BTW, It supports IronOS and devs got the scehmatics for it: https://github.com/Ralim/IronOS/issues/2010
>>2997361>BTW, It supports IronOS and devs got the scehmatics for it: https://github.com/Ralim/IronOS/issues/2010Oh huh, didn't see that on the IronOS homepage a year ago.Looking now, my Si012 Pro has the same PD controller IC and same MCU, just the board is the V1.4 not the V1.5. Its MOSFET IC can handle 4.9A which is more than enough, I may try to flash IronOS onto it using that .hex file, and get an ST tip to test with. Having a bit more grunt could be handy. If that bricks or fries it, then all the more reason to buy a Pro Max.Additionally, that Pinecil page says the following:>Older Pinecil V1 can not use the short tips until firmware code is written to enable manual selection of 6.2 Ω or 8.0 Ω tip>Only the V2 model has the hardware to auto-detect the two kinds of tips short 6.2 Ω or regular length 8.0 Ω>If you would like to help with the code, see GitHub/IronOSSo it seems possible that the Pro Max is equipped to measure tip resistance while the Pro is not.
>>2996960updooted>25 BOM items>28.5x2.3 mm>castellated "fork" for soldering the coax shield tothe one complication besides the edge plating is that I need the vias filled. but at least I got away with not having to put those special vias-in-BGA-pads
>>2997402>2.3 mma wien bridge oscillator that fits in my pee-hole?i'll take a dozen, pls
A couple of horrible bodge wires later and my board works. But I wonder if I am setting myself up for a gradual failure over time, so would appreciate some advice.I am converting a sawtooth wave to a pulse wave with a comparator, so the op-amp is swinging between max and min voltages, -12 and +12. Elsewhere in the circuit, I want to send only the positive half of that pulse wave through an analog switch, which can't handle the negative part (it is powered between +12 V and GND).So after buffering the pulse wave at U1C I pass it through diode D1. This gives me the expected positive-only square wave for the analog switch and for a flip-flop frequency divider. But when the output of U1C goes to -12V, will I develop a slightly negative voltage at the GND side of D1 due to the diode drop voltage? Could this damage the flip flop and switch by exposing them to a slightly negative voltage?Please and thank you anons, I am so close to completing this project.
>>2997471I was so close to getting it right...thank you to the anon that recommended that desoldering alloy, it allowed me to easily remove the CD4053B that I cooked by powering from +5 volts while it received +12V input signals
>>2997471That diode won’t have enough leakage to cause the voltage to go significantly negative. Maybe a few mV, it’s only an issue once you get into the realm of 0.7V, and even then it’s current limited.Also shoulda used LM311 comparators for their intrinsic speed and level shifting capability. TL074s have ok skew rate for an op-amp, but they’re just not designed to be used with positive feedback.
>>2997475In a previous design I was using an LM393 comparator to get the pulse wave from the saw. I can't 100% remember why I removed it, but I think it was more convenient to use the TL074 op amps because there are plenty else on the rest of the board so it saves having a separate kind of dedicated comparator IC, which I would have to waste half of.This is just for audio frequencies up to about 2.2 kHz, so even if the TL074 isn't ideally suited, I can still get away with using it, right?
>>2997479Depends more on how fast you want that edge to be, but it’s probably fine. You might be able to add diodes to stop it from saturating itself too much.
I have a USB powered fan and I want to put in a speed control for it. Do I just wire in a pot in series? What values? Logarithmic or linear?
>>2997543You should use a PWM controller.
>>2997509>stop it from saturating itself too muchIs that a problem?I am going off of textbook understanding and simulations, but I know that doesn't necessarily capture every real world consideration.
>>2997546I've been doing some more reading and it looks like that design is FUCKED UP and I should use a proper comparator instead
>>2997543>>2997545I know this is /ohm/, but you can get dirt cheap USB PWM controllers from China all day long. There's a common "fan governor" with a pot on the side too, but that one just varies the voltage from 2.5-8V.
>>2997546It’s not going to damage anything, it would just make the response time of the op-amps a bit faster, maybe.
>>2997048I want 88MHz to 108MHz transmitting stereo audio.I'm making a portable music player. The intent is to use the RP2040. I plan to have a bluetooth transmitter as well.I found this pre-built circuit that does exactly what I want. Is there any reason I shouldn't use it?https://www.alibaba.com/product-detail/VMR6512-Hi-Fi-FM-Transmitter-Integrated_1601591382662.htmlWith this module I can even have the radio transmitting different audio tracks as the bluetooth and headphone port. Could even have the USB-C port outputting a different audio too.I can simplify the design and make my own transmitter later on. I just don't have the understanding of frequency modulation to do it right right now.
>>2997604 (me)There's also thishttps://ja-bots.com/wp-content/uploads/2026/01/KT0803L_V1.3.pdf
>>2997604Both those modules look good. I’d go for the first one (not necessarily from alibaba) and feed it digital audio, as that removes an unneeded DAC->ADC conversion.FYI, your MCU of choice may struggle with some audio formats, both for speed and RAM. Lossless and MP4/AAC especially, but even MP3 may pose a challenge. MP3 players from 10 years ago likely have more power than a 2040, at least in the narrow domain of audio processing.
>>2997546 (me)>>2997552 (me)I breadboarded an alternative design for my saw -> square wave with an LM393 comparator and it works. This keeps everything in the positive domain, don't need to worry about the diode, etc, and the edges seem sharp enough to drive the flip flop as a frequency divider.This should be a lot better, right? I honestly can't remember how I convinced myself this was not the way to do it. I'm sure I was previoiusly running into problems with the sharpness of the edges.
>>2997624Simulators are a plight on education. They invite people not to think, not to learn but instead just 'simulate' something without understanding how the simulator works, what limitations result from the inner workings and quite frankly without an idea what 'question' they are even 'asking' the simulator.This simulator is one of the worse offenders, as it seems to integrate several elements of the equivalent circuit of a real part into a symbol that is commonly associated with the ideal part.To begin with: What do you expect to happen at a node that is connected to a fixed voltage source and a (presumably voltage) signal source?You will find more value in reading about the basics of electrical engineering first and then look at circuits that are commonly employed to achieve what you want, conduct your analysis using oen and paper. You might find more value in analysing circuits made up if discretes only first, before checking how ICs might be employed. Once you understand the underlying principles and you have very specific questions regarding performance, stability, losses etc. then use a SPICE simulator and model relevant parts of your circuit.
>>2997626Learn how transistors work, but only to interpret the input and output sections of IC internal equivalent schematics. Op-amps are a more ideal and beginner friendly way to learn about analog circuitry. Learning about proper biasing and impedance problems can wait a little.
>>2997614Thanks mate. I expect the first iteration will be more for me to figure out how to make it right. I'm not too worried about getting it all right now. I'm mainly using the rp2040 because I'm familiar with them and have them on hand. Likely I'll change to the ESP32 or something and that should allow me to simplify a lot on the device
>>2997543Honestly a pot might work for a little DC fan. So long as the fan is sufficiently low powered such that the pot doesn't get roasted. Takes math to calculate the resistance though, and it depends on the current draw of the fan. Be easier to have a voltage-dividing potentiometer + an emitter follower, or a current-limiting potentiometer + a current amplifying BJT.
>>2997653>Honestly a pot might work for a little DC fannopeordinary pots are rated 1/8-th or 1/10-th wattno fan is so weak that a pot would survive unless it was wire-wound, or packed in dry ice, or held under a witch's tit
>>2996880Hi,I'm relatively new to electronics. I've been doing security research for a decade and only in the past year had an opportunity to work on a project and develop attacks on a HW piece of tech. I got myself a logic analyzer and learned how to use it and got good at it. Anyway, I have some time off this summer and would like to build my own electronics "lab". I have a soldering iron (crappy one) and a Saleae logic analyzer. That's all. At work I had an oscilloscope, DMM, microscope etc.Since the OP post has no actual equipment recommendations (a big omission IMO), maybe someone here can recommend me some tools. I'm not interested in bottom-of-the barrel stuff but also not the expensive shit that only the corporations can buy. Can someone recommend me a:-DMM-Oscilloscope-Soldering station-Hot air station-Microscope-Bench power supply-Hand tools: tweezers/screwdrivers/pliers-Wire strippers-Board holderAnything else I'm missing???Thanks in advance!
>>2997626I think you are hinting at the fact that I was missing R5? I chose R76, R75, R5 so that when the LM393 output goes high impedance I can't develop more than +5 volts at pin 3.
>>2997665There are so many products available in different regions it's hard to recommend anything specific. Instead of buying a piece of gear that you think you might need, just buy gear to solve problem you're working on. It's pretty useless to, for example, buy a 30V 30A power supply just to find out that output switching noise is messing up your sensitive circuit. Price is also sometimes not a good indicator of product quality. Sometimes cheaper option works better in specific applications than expensive one.
>first attempt: shit's broken>attempt to diagnose, isolate subcircuit>desolder IC and find one completely shot>/ohm/ points stuff I overlooked>squeeze in a few resistors>resolder everything>virtual ground still pulled down to the negative rail by *something*I want to fix this and get it working so bad, but fuck me I'm losing my sanity here. Wish me luck.
>>2997658Worst case it drops half the power. If it’s a little 30mm fan it’s fine, but a 120mm definitely not.>>2997675That works? Doesn’t the LM393 pull its output hard down to your negative rail?Also I’d be using a Zener to ensure its voltage remains under 5V instead of resistor tuning, as that allows me to more easily tune the other resistors.>>2997729Did you breadboard it beforehand?
>>2997731>That works? Doesn’t the LM393 pull its output hard down to your negative rail?Yes, but the LM393 is connected between +12 and ground, so it gives me a square wave at positive voltages only that the flip flop can handle (which is also wired between +12 and ground).Measuring the point between "PWM" (the DAC input) and R77 on the breadboard only ever shows me the voltage at which the DAC output is set, so I think I will be safe.
>>2997700>Instead of buying a piece of gear that you think you might need, just buy gear to solve problem you're working on.Good advice but not in this case. I can't do much without a DMM, oscilloscope, soldering iron etc.
>>2997746Well you could start off by explaining what you’ll be using the tools for. Speed and memory of the scope, power and size and portability of the iron, specs of the DMM, they depend on what you need to do with them. I’ve got two of each of those with different use-cases.
>>2997731>Did you breadboard itYes, of course, except for all the bypass caps, that were suggested by /ohm/ as I was about to solder, so I quickly threw them in. Also, some subcircuits were breadboarded separately for clarity. The only difference I had was the virtual ground being provided by an opamp on the breadboard, whereas the perfboard version has a simple voltage divider. Thing is, the virtual ground is hooked only to positive and negative through the bypass caps, and to opamps/comparators' inputs, so I'd expect it to not drift too far off target even with a simple divider due to the almost-zero current it would end up conducting. At this point either another opamp is shot (idk how they could die just like that), or I fucked up the schematics.
>>2997753You’ve triple checked your pinouts? Yeah all I can suggest is socketing your op-amps so you can swap them if needed.
>>2996880why is this dumb bitch pointing at resistor pads when they're carrying a cart full of capacitors? there's two pads right there that need caps.fucking dumbass normie artists
>>2997788Follow her pointing arm in her shadow, it's pointed downwards but only shallowly. The target of her pointing is off-screen, possibly a cap connected to U7's 7th pin. C50 and C51 aren't necessarily the same size capacitors as C49.
I managed to leave a nasty scorch mark on my desk. I need a soldering iron stand.Is this a good one? What do you guys have?
>>2997820>What do you guys have?i like this one coz it leaves the hot bits completely uncoveredany faggits hanging around my workstation will burn themselves, their pants or their expensive winter coats as they peer into my workteaches 'em to stay the feck away
>>2997822Kek, nice. I live alone so I have diff kinds of issues. Also, my desk is wobbly so I'd like something firmer. Also, how do you clean the tip on that little sponge? I;d prefer a bigger sponge so it lasts longer.
>>2997788>triple dubs wasted on low tier misogyny>against 2D no lessI hope you get better anon. It took me a long while, but if I can do it anyone can.>>2997820Pic related is great. Very stable, and using the brass sponge won't thermally shock the iron tip.
>>2997827>Pic related is great. Very stable, I like it! Any idea where to find one of these things?>and using the brass sponge won't thermally shock the iron tip.I had no idea about this. I've been using a wet sponge for like 2 years. I did some searching and you're right. i've been doing it all wrong.well, I can't find any iron stand without a sponge. since sponge is actually bad, I don't want a compartment for it at all. I did find this but you need a 3D printer for it.
>>2997845You should still use a sponge to clean the tip when you finish working. Then tin it with fresh solder and turn it off.
>>2997845The one he posted is specifically designed to work with T12 soldering stations, some irons, especially non-temp-regulated irons, may be too thick to fit.I’ve used copper wool and wet sponge in the past, I don’t really notice a difference in tip longevity. Wet sponge seems to leave the tip cleaner, though my copper wool probably needs replacing.
>>2997845I got it on banggood when I got the whole setup a ksger T12 and ksger SMD station i.e. the hot air gun, but it was a not insignificant number of years ago. It was all cheap, but it all works (mostly: I had to isolate the ground of the hot air because the heating resistance expands and shorts with the metal jacket which is grounded). The brass sponge holder was sold separately, and I did glue a rubber mat underneath the base, for added grip and reduced surface wear.
Is there a good list of component abbreviations for printing on PCBs? Like R for resistor, C for capacitor, etc.?
>>2997892https://resources.pcb.cadence.com/blog/2023-identifying-electronic-components-on-a-circuit-boardbut best to follow up with an AI query coz it's more messy than thatdiff people use diff letters, and there's often 2 letters for the same componente.g. a chip could be an IC or a U
>>2997892https://en.wikipedia.org/wiki/Reference_designator>>2997912And a voltage regulator could be IC, U, or it could be VR. But I've also seen VR used for variable resistors. There are formal lists, but they're not always adhered to.
My Sugon A9 arrived. It uses T210 tips. These things are ridiculously small. LOL. When you see pics online or the reviews, you have no idea about the actual size. They have no thermal mass whatsoever. Yeah, don't expect this size to be a general iron. T245 is probably a better choice if you need that.
>>2997665Ok, I'll give it a try, so at least you've got a starting point>-DMMFluke is usually the corporate way to go. And for a reason. They are reliable and good but you'll also pay for the brand. Personally I prefer Brymen. You'll pay half the price for 2 times the features. I've got a BM867s and I'm happy with it.>-Oscilloscopeno cheap recommondations on that part. Good stuff from work for the price tag of a new car is probably too expensive.Personally I've got a Siglent SDS 1104X-E. It is decent but I did not push it far, so who knows. You can hack it though:https://www.eevblog.com/forum/testgear/unlocking-siglent-sds1104x-e-step-by-step/msg3344060/#msg3344060>-Soldering stationI've got myself a "cheap" JBC station. And I don't want to use anything else. Used an old Weller at work and switching tips is just a pain in the ass.>-Hot air stationPersonally I don't have one and only ever used it at work. Don't know the brand but it was pretty old anyway. JBC makes the too. If you want to sell your kidney.>-MicroscopeI don't use one. I've got myself a baseball cap and ghetto-rigged a bracket for different magnifying glasses to it.>-Bench power supplyI've got a single channel and a dual channel digital power supply from Korad. Fairly cheap and I had no problem with them so far.>-Hand tools: tweezers/screwdrivers/pliersKnipex for pliers, Wera for screwdrivers. And I've got a chink no name repair kit for smart phones.>-Wire strippers95% of the time I use a pair of side cutters. Just get good.>-Board holderI think I've never used any professinal dedicated board holder>Anything else I'm missing???If you are working with anything that's connected to mains, so 110V 220V 230V 400V etc. (if you are not a professional you shouldn't) a isolating transformer can safe your life.A desoldering pump. Some use a big ass station. I feel very comfortable with using the small syringe like thing. And it costs almost nothing.Hope this helps
>>2997980To be fair T12 tips are stupid long. What’s the beefiest tip you can get for a T210? Is there like a D4? I’m considering getting one. I saw what looked like a D4 in the “C210-007” size, but the legend said it was 2.3mm not 4mm, man that scale messes with you. I’m going 210 because that’s what battery irons are using, not because I want to work on fine things.
>>2998003>To be fair T12 tips are stupid long. I'm not sure what they were thinking. Probably to give it some mass for thermal inertia?>What’s the beefiest tip you can get for a T210? I think it's K. It's like 2mm wide.>Is there like a D4? No. There's like I, IS, K, KU. That's it.?I’m considering getting one. I saw what looked like a D4 in the “C210-007” size, but the legend said it was 2.3mm not 4mm, man that scale messes with you. Yes, when you see these tips on YT or in images, they look huge. They're so damn small.Here's a comparison... it's much smaller than a ballpoint pen. T115 is even smaller... like 2/3 of the size of T210.>I’m going 210 because that’s what battery irons are using, not because I want to work on fine things.That's a silly reason. It will sit on your shelf collecting dust. You won't be able to remove even a through hole capacitor.I got the new station for phone repair. I've been buying broken phones and fixing them and selling them. I needed something for fine pitch SMT.
>>2998014>I'm not sure what they were thinkingAllows it to anchor in the base of the handle solidly, while retaining a long but sufficiently strong stainless shell for high thermal resistance from the hot element down to the electrical contacts would be my guess.>Probably to give it some mass for thermal inertiaI don't think so, only the last 1-2cm of the main cylinder are solid with the element, the rest is hollow. Any iron relying on a long thin thermal mass would be worse than a fat wide thermal mass like the C470s.>I think it's K. It's like 2mm wide.I'm seeing listings with K tips that are 3.4mm wide. From JBC themselves are the extra fat C210-039 and -042, which are 2.9mm and 3.9mm wide chisel tips respectively. Same for the -040 and -041 bevel tips. Though I've always wanted to try a barrel tip for THT work, that's got a whole bunch more surface area for both the lead and pad than even a chisel tip, though your work angle is fixed like with a bevel. Strange that T12 and TS and such tips have no barrel or cup tip varieties.>That's a silly reason. It will sit on your shelf collecting dust. You won't be able to remove even a through hole capacitorThermal mass aside, these irons can put out 65W. Might take a bit of time with big ground planes or other cases where an instantaneous 65W isn't sufficient, but so long as the thermal resistance from the heating element to the tip is low (it's a cartridge tip so it should be low) that's more important than thermal mass.Also the point is that it's for a battery powered iron, you'd leave it in your toolbox, not on your shelf.
>>2998019>which are 2.9mm and 3.9mm wide chisel tips respectivelyDiagonally. It's much less by width (between 2 and 2.5mm).>these irons can put out 65WMy A9 can put out 120W but that will burn up tips quick. These things are meant to operate at 40-50W max. You're supposed to use it for nano soldering of SMT components. Mostly re-work.
>>2998035>DiagonallyNo I’m talking about the chisel tips, not the knife tips. Pic related from JBC’s website.>My A9 can put out 120W but that will burn up tips quickWill it? There’s a thermistor right next to the heating element, so I’d never expect it to overshoot enough to be damaged by overheating the element.
>>2998036>from JBC’s websiteThat tip costs as much as most irons from China. Chinese haven't cloned that tip so it's pointless to me. If I had money to burn, I'd blow it on women and not overpriced tips. It's much more economical to buy a whole iron that's suited for the task than trying to get some extra juice out of an iron made for nano soldering.
>>2998042>That tip costs as much as most irons from ChinaYeah that's the other thing. The chinky T245 tip variety looks much greater than their T210 variety. Like spatula, barrel, spoon, fork, etc. Even solder pots and heat-set insert tips.>an iron made for nano solderingIdk, the thing puts out the same amount of power as irons with T12 and other tip categories:https://www.aliexpress.com/item/1005012335736983.htmlI just wish someone made a high-power T245 battery-powered iron. For 65W out of a lipo at 100C discharge rate that's just 170mAh from 1S, so it's definitely feasible to get more than 10-30W out of a battery iron. 3-20Wh is more sensible. If it draws 10W standby to stay up to temp, then that's 20 minutes to 2 hours of standby use. Though you might ditch tip longevity and just cool it down and heat it up as needed since these cartridge tips do that pretty fast.
>>2997988Thank you anon for an awesome list!>BM867sI really like it. I was also told that BM235 is quite good. Not sure which of the two is better. I'd probably go for a cheaper one and put the 'savings' into an oscilloscope.For the Fluke, after looking at a bunch of specs, I've narrowed the choices down to 116 or 117. I'll see if I can find any of these models used and if not, I'll probably go for the cheaper Brymen .>Siglent SDS 1104X-EOh, that's a nice one! I'll have to take a closer look. For whatever reason, I fixated on a Rigol DHO814. It checks off most of the things I need and like.At work I have access to a Tektronix MDO and I've learned how to use an oscilloscope on that. But those things cost more than my car.>I've got myself a "cheap" JBC station.I'm gravitating towards that as well.I have a lead on a used CD-1SQF at an industrial auction and I might make a bid for it.>KoradNever heard of them. But I do like the prices! Thanks.>95% of the time I use a pair of side cutters. Just get good.We had a pair of some German-made strippers but the name was rubbed off. I liked them. Would like to buy a pair because nicking wires gets boring quick.>isolating transformer can safe your lifeMost of the stuff I'm working on is small and battery powered but isolating transformers are cheap and good to have. I'll add it to the list.I think I can build-out a good home lab for under $2k.Thanks again! If anyone has other suggestions, please let me know.
>>2998074>We had a pair of some German-made strippers but the name was rubbed off.If the brand color was yellow I would take a bet that it was a Jokari.They got a collection of wire strippers. One for every taste.If it was red/blue it was most likely Knipex. Great pliers but they also got wire strippers.
Tested these LEDs at 3.3V & 1kOhm resistor to see if I could use them in my game project as indicators. Current was about 1.4-1.6mA for all of them. All the reds are quite usable. They could use even larger resistors (maybe 1.5kOhm) to make them more even with the other colours. Clear yellow & green are okay too. The frosted yellows & greens are unusably dim even when viewed from the front. The green side on the two bipolar LEDs is a bit dim too. Maybe could try that one with 680R or something.
>>2998090Side view. I plan to ditch the frosted 5mm yellow&green and try the others with: - reds: 2.2k - yellow: 1.5k - green 5mm: 820R- green 3mm: 1k - bipolar red side: 1k- bipolar green side: 680R
>>2998092>I plan to ditch the frosted 5mm yellow&greenand yet, they are the fairest of hue, texture and coherencetruly high-borne princes of the landif they need a bit of extra current, you must give them their due, to thank them for their gift of harmony and beauty
>>2998090You could always run them from the battery or other upstream power source, but that will create a brightness variation with battery capacity if you use resistors, and I doubt you want to voltage mirror them. If you have a 5V rail I’d definitely use that.Also that yellow clear LED looks gnarly. Full of bubbles.
This is the first thing I've ever built and I have a bit of an issue, I've ordered my pcbs already but d9 is a rats nest wire, it should be fine if connected it to the ring on the right with a piece of cable, you can kind of see what I mean on the pic, but the thing is so small I don't think I'll really be able to do it, could I like just scratch a line with a knife to connect them or is that a no go, or are there any other alternatives? I can post a pic of the pcb if that would help.
>>2998177Are you saying the trace from D9 wasn't in the design but you need to add it? If so I'd sooner add it a bit further down, as to join to the ground plane instead of the pad. Using a knife or whatever to scratch away the solder mask. Just makes it less of a pain to handle when you solder that ground pad.In future, run the "design rules checker" or DRC in your eecad software of choice, it should show all unconnected components, overlapping footprints, and other errors.
>>2998178That's what I meant yeah I need to add a trace, going diagonally is not a bad idea I'll try that, should I cover the exposed copper with something after doing it, wouldn't it oxidize and cause issues later down the line? And you're right I really need to step up my game and check designs before I order them. Here's a pic of the pcb, my glasses are on the left for scale.
>>2998179>wouldn't it oxidize and cause issues later down the line?It should be covered in solder, wouldn't matter for oxidization purposes. But it might matter for rigidity of the wire itself. The distance is short so it's probably fine, but some would lock the wire in place with a dab of UV-cure glue/resin/enamel, or solder-mask.The DRC also yells at you for overlapping silkscreens. Pretty complex for a first board, what does it do? Hope you got as many components in the basic/preferred parts library as possible, assuming JLC.
>>2998181Yeah I got it from JLC, I had to hand pick a few components but in the end I got pretty lucky.>looks complexI didn't design it, I don't have enough knowledge as of right now, it's something I found on the internet and thought it looked interesting so I wanted to try and put it together, it's for electrolysis, I still have to do the bottom side by hand that'll make for some nice soldering practice. Thanks for the help too I'll go about fixing it once the rest of the components come in.
>>2996884bingo. Wtf OP did you even have to say that??Also you could run electricity through it and make circuits.Wainscoting Herringbone
>electrolysisHad a customer at work wanting to buy a rheostat to vary power, 12V, up to 50A. After some interrogation the mentioned the load was sorta like electroplating, and the control needed to be varied through a mechanical linkage, and it clicked that he was making an HHO generator inside a car. He kept dismissing my suggestion of a DC-DC converter, even after I showed rheostats online were like $300 minimum. I didn’t try convincing him that it was a negative-sum process or might cause predetonation, far too stubborn for that.
> https://toptechboy.com/lesson-26-wire-wrapping/wire-wrapping/A long time ago somebody gave me this exact tool, but I had no idea what it was. I was like, cool a screwdriver with no end. thanks, buddy. They were like, "here kid", and left without doesn't explaining anything. The other day, I saw someone comment about using a wire wrapping tool instead of soldering. I thought, wot is a wire wrapping tool? And so I solved the ~decade old mystery. Now, your telling me all this time I could have been wire wrapping with this tool at the bottom of my toolbox instead of investing money in soldering tools and huffing solder fumes???thanks for reading my blog.posting a link cause my IP range is shit listed and I can't post images.
>>2998353https://learn.sparkfun.com/tutorials/working-with-wire/how-to-use-a-wire-wrap-tool>have a breadboard >still use a tool to twist wires to component leads point-to-pointjust toss it
I need my microcontroller to measure the frequency of a +/- 12V square wave, will this work to shift the logic level? I think that I can't just apply the +12V to the transistor base.The transistor will pull down a 3.3V pin that has an internal pull-up resistor.
>>2998562I don’t see why the first one wouldn’t work, or why Vbe would be a problem when you’re running 12v into it. The more classic approach would be to replace the series diode with a diode from emitter to base. Also 4.7k on the base with 50k on the collector is kinda silly, I’d use 100k on the base and 10k on the collector. Though if the MCU has a built-in pull-up, I don’t see why you’d need a collector resistor at all.If you’re extra paranoid, or might have issues with a ground rail shared between the square wave and MCU, you could use an optocoupler.Using a small-signal MOSFET like the 2N7000/7002 is also an option, they’re more than tolerant of +/-12V on their gate if I recall. A gate resistor is still not a bad idea, and depending on your square wave source and FET, a TVS diode and/or gate-to-ground resistor, since those gates can be ESD sensitive.
>>2998582I think I didn't mean VBe, on further thought, I'd misunderstood some other explanation I'd read. I guess I am wary because I've so far only generally seen transistors to shift to a higher voltage level, not using a high voltage to control a lower one. Usually I manage by repurposing some other schematic I found online but couldn't quite find one that matches this application.The idea with the diode is that I should only get the positive parts of the square wave to control the transistor base. So pic related would be better? I guess when the square wave goes to -12 V, the diode makes it so that the base is at ground potential?
>>2998585
>>2998591I do actually have some optocouplers to hand and am coming round to the idea of using one. Isn't it over-complicating things, though?
>>2998593Yes and no. It will work safely though.
>>2998591a typical opto-coupler's LED reverse voltage rating is 6Vso a series diode on the input is still a good idea
>>2998585>the diode makes it so that the base is at ground potential?Well more accurately it goes to -0.7V, but that's well within tolerance. Either way the current through the diode and base-emitter junction (which also acts like a diode) are being limited by that resistor. Actually it might be fine without the extra diode, I can't find any reference in a few BJT datasheets of absolute maximum reverse base-emitter voltages. Though I'd err on the side of caution.Not quite sure whether you mean a GPIO or a supply pin when you write "3.3V MCU pin". If the former, your resistor placement is wrong, pic related. Either you have an external pullup resistor to the same voltage rail as your MCU, or you enable the MCU's internal pullup resistor. If the latter, use proper symbols for voltage rails.
>>2998620You're right, but maybe an anti-parallel diode (to the opto input) would be better, like >>2998633depicts.
>>2998660Or just use an AC opto.
My boards came in today. My diy rig grows today.
battery-powered soldering irons based on vape handle are under $10. Has anyone tried them yet?
Newbie here. How do I measure the amount of watthours that my esp32 + sensors will consume? I need to know what kind of battery to get. There are documents that estimate the range of amps but I want to actually measure it.
>>2998722plug your power supply into a kill-a-watt
>>2998723I have a multimeter. Can I use that somehow? Or maybe the esp32 can run a script that keeps track of the power used? Idk. Is it worth spending $50 on a separate device?
>>2998725Sure you can, but there is a burden voltage from a multimeter that can cause voltage drops. With something like an ESP32 that can cause it to crash and have issues with how finicky the 3.3v is.Buy a KWS-2303C and itll tell you what the draw from USB-c is and it should be pretty close.
>>2998725Can you use a 5V power bank?
>>2998723>>2998728>>2998729Okay I can see it's worth having a specialty device. These are all good options, thanks. I'll probably go with kill-a-watt since a standard wall outlet has more uses than usb-c. I can play with it to measure other devices in my home at least.>>2998729That would be smart to just measure how quickly it drains a power bank. Unfortunately I don't have one. I wanted to measure first so I knew what size battery to buy.
How do you boost IR reception? I have a desk fan that is controlled by a remote. It was always bear to get it to work so I eventually got a programable remote and cloned the original. This remote is much more powerful (I can control other devices from much greater distances), yet my fucking fan is still a bear to control with it. I cleaned the IR receiver window but it didn't help.
>>2998661>>2998660>>2998633>>2998620>>2998594Alright I'm hoping anons can help me over the finish line given the amount of time it took to draw this. It all works with the optocoupler. I added D1 to protect the optocoupler LED from negative voltages. But the input +/-12V square wave is generated by a comparator with an open drain output. It feels like a problem that there is nothing to limit the current through D1 when the LM393 output is at -12 V. Do I need a current limiting resistor? I can't put too much resistance between +12 and the LED because the optocoupler needs about 10 mA.
I want to start with electronics. What's something I can do with aside from leds?
>>2998765>I want to start with electronics. What's something I can do with aside from leds?buy a cheap kit to learn how to solder. it's one of the most essential skills. when choosing kits, pick something that doesn't have any ICs and is analog because you won't learn anything if you're just soldering a bunch fo ICs. something like pic related is great since it uses a single transistor and you can actually understand how it works.you could also go down the breadboarding route but you won't learn how to solder and you'll be overwhelmed by the choices. If you're a dedicated self-starter, that might be a better option. But it will cost a lot more.
>>2998722I like the little USB power meters, though if you’re not powering via 5V with a battery but going into an integrated battery input on your MCU board, it might differ substantially from the 5V measurement due to the quiescent current an inefficiency of the regulator. Only usable method is to have something add up the amp-hours directly going into that battery connector.The next best option would be to measure the static mA drawn when it’s woken up and going tasks, extending those tasks if necessary to get across multiple DMM samples. Then doing the same thing to measure the μA drawn in sleep mode. And then calculating the sleep and wake time and adding it all up. Maybe doing it at 4.2, 3.7, and 3.2V to see how the power draw changes with battery voltage.If your code is already done, you could just use an 18650 or other spare lithium cell to test it. Power banks turn off when current drops too low, and have their own quiescent current you don’t want to deal with.The better option is probably to use another arduino or ESP or whatever with its ADC to measure voltage and current continuously and spit them out over USB UART over an hour. Cheap and accurate, I like to use a length of thin magnet wire as a current sense resistor, you’d just need to whip up a little current sense amplifier, LM358s can sense to the ground rail.You are using deep sleeps to extend your battery life, right anon?>>2998723With this you’ll get all of the inefficiencies of the wall adapter. For a 1A load maybe thats 5%. For a 50mA load the losses will probably be greater than the quiescent power draw, and for a microamp sleep load it’s thoroughly unsuitable.>>2998739Might be a slightly different IR wavelength?
2998762Usually you’d have the comparator pull current through the LED, not rely on the pull-up resistor. If the output polarity is wrong, invert the comparator’s input configuration. To avoid the diode, just put the LED’s cathode at -12V and use an appropriate resistor to limit current. Furthermore, if you want a nice square edge, you’d put the comparator on the output of the optocoupler, not the input.What’s the signal source in the first place? You probably don’t need the opto.>>2998765Why do you want to learn electronics? To make robots, synthesisers, radio transceivers, IOT devices? Or to fix or reverse engineer hardware? To get a job? Do you want to build other people’s designs, or make your own? Do you want to outsource production, or get your hands dirty? Only once you’ve set a goal can you work on your roadmap.There are general skills worth understanding for any field, like the basic intuition behind ohm’s law and equivalent sources, input and output impedances, and common components like diodes and capacitors.
>>2998769>Might be a slightly different IR wavelength?I mean, the original remote has the problem as well as the universal remote. The universal was able to read the signals from the original remote no problem. The only thing having an issue is the fan itself.
>>2998770>Usually you’d have the comparator pull current through the LED, not rely on the pull-up resistor.That actually makes so much more sense and solves the reverse polarity problem, thanks anon.I think I could well use a transistor instead of an optocoupler like >>2998633 but it seemed wise to keep my 3.3 V Pi Pico separate from the rest of the circuit which is +/- 12 V. The comparator is required in the first place to work as a kind of zero crossing detector so that I can measure the frequency of arbitrary input waves.
>>2998773Right, makes sense. To speed up the edge of the optocoupler, you'll want to mess about with the LED and phototransistor resistor values. The datasheet might say something about that, usually in the context of a current gain, e.g. 1mA into the LED corresponds to 500uA in the collector. Too large of a phototransistor resistor and it will react too slowly, but too small and it won't saturate properly, giving you something like 3.3V max and 2V minimum, for example. No clue about that 1M resistor, never seen aything like that myself.
What's the best way to make a housing for a DIY linux handheld without a 3d printer
>>2998739>How do you boost IR reception?if it bothers you so much, you buy or make an IR repeaterthey come in wired and wireless flavorsalternately, you walk up to it and hit the switch with your cock
>>2998809>without a 3d printeruse a 3d printing service like jlc3dp.comto get it made fast, go local: google "3d printing service <name of my town>"
>>2998809Designing your parts in CAD and paying someone else to 3D print or CNC machine them. JLC3DP are pretty good, though there might be local options with faster turnaround if you need that rapid protyping ability. Depending on what and how much you're ordering, it may be cheaper to buy the printer and be done with it, but it may understandably be something you don't want to waste your space or time. Especially if you go for resin parts.What kind of hardware are you going for? CNC machined aluminium is attractive for its heat sinking ability, but depending on CPU choice that may not be an issue. Just as important as designing a good case is designing something manufacturable, with 3D printing it matters somewhat (less so with SLS) but with CNC machined parts it really matters. I wouldn't really trust the chinks with designs that need to be reoriented by hand, there's room for error in the alignment (I've seen people run into problems like this from chinese CNC shops) and it's also probably more expensive too. A lot of my PCB part selection decisions come down to soldering price.If you're obtuse, you can try to go analogue. Make a mould or plug using modelling clay or plaster of paris, then use that to make a fibreglass or CFRP hard shell case. No CAD needed, but probably a silly idea.
>>2998809>linux handheldwat
>>2998821>watpoor deprived kid, your mom never got you one?
>>2998762>Do I need a current limiting resistor?Maybe between D1 and ground, so it doesn't create a voltage divider with the 1k resistor. Just guessing.
>>2998762>Do I need a current limiting resistor?put the diode in series with the LED and good as goldor at least pyritethe idea of putting a diode to ground makes sense for transistors, coz you're protecting against electro-magnetic noise that can turn on transistorbut LEDs are 217x less susceptible to EMPs so you dont need that
Optocoupler is working in a much simpler setup now, thanks to everyone who helped.>>2998777>No clue about that 1M resistor, never seen aything like that myself.I heard that it is used to "remove excess charge that builds up at the phototransistor base". It seems to improve my rising edge times. Although to actually measure the frequency on the MCU I will use the falling edge, which is a lot sharper. I can get a faster rising edge with a 1k pullup resistor, but that seems rather small for that purpose??It does seem a bit wasteful that the optocoupler uses ~10 mA (or I suppose 5 mA assuming a 50% duty cycle), but that's from the data sheet test conditions.
>>2998911converting a bipolar sine wave to a positive-only square wave :^)now I just need to write the frequency measuring code
>>2998911Wow that 1M does make a difference, I'll have to keep that in mind, shame my discrete phototransistors only have two pins.The datasheet says it has a current transfer ratio of at least 100%, so I'd try to design it with the same current through the LED and phototransistor. If they were at the same voltage then the same resistance would be about right, but because your LED side is on 24V and your phototransistor side is on 3.3V, your phototransistor side resistor should be approaching a much smaller value for optimum speed. Like 290 ohms, though I suspect that's diminishing returns. You're currently giving the LED side 10mA, the opto can handle up to 50mA which might be a tad faster, but I'd leave it as is. Maybe try a ~200k resistor instead of 1M if you're curious.>>2998916There's a little bit of phase offset, is that the maximum frequency you care about?What's this all for anyhow?
>>2998918300 ohms on the phototransisor side gave me a 5 us rise time, that seems bretty gud. Thanks for pointing out the need to match the currents on either side.I don't think the phase offset will affect my application. This is self-calibration circuitry for a synthesizer - I need to calibrate multiple voltage controlled filters to have the same response curve. So I'll measure the self-resonant frequency at a couple of different input voltages to establish calibration curves for each one, so I can get them in tune with each other.Rather than reference the comparator to ground, the next part of my plan is to use a DAC output as the reference instead. This should let me measure the maximum amplitude of the signal by increasing the reference voltage until a square wave is no longer generated.Been working on this for about 7 months on and off...
>>2998928Neat idea. I guess you'll use an analog switch IC (e.g. CD4051) to select which filters and waveforms to measure? Optical isolation isn't a necessity inside a synth, I'd probably have just gone with a resistor network to level-shift the signal, followed by clipping/TVS diodes instead for input protection. That would allow you to use the MCU's ADC or analog comparator directly.
>>2998930I guess there are many different ways to do it. Surely the optocoupler adds a bit of safety by isolating the MCU from the much higher voltage analog stuff. I've already accidentally killed one output pin by exposing it to the wrong voltage
Is it even possible to sodder these DFN-style SMT LEDs with an iron? Or should I just stick to something like 1206 or THT LEDs?
>>2998991>with an irononly if you want to experience pain.a hot plate or hot air is what I'd use.
I bought a set of trimpots on aliexpress (among other things)The outside bag was full of a clear bubbly liquid, and each smaller bag with the actual trimpots have the same liquid in them.Almost looks like soapy water, has no real smell that I can tell.Everything else in the package was dry, just the trimpots.Have you ever had this happen before? Am I going to die from exposure to this extremely toxic liquid?Is it going to ruin the variable resistors?
>>2998991>>2999006Looks doable with solder paste and an iron. Add a fat via under the each pad so you can reflow them with an iron from the underside if you want to try that.>>2999009Could be something like a silicone dielectric grease or oil.
>>2999009>Am I going to die from exposurenever heard a more outrageous story in this domaini would junk it immediatelybut, if you're curious,. go back to the product page and read the reviewsthere may be an explanation from a customer
>>2999016Its fentanyl juice, touching it will kill you
>>2999015>Looks doable with solder paste and an iron. Add a fat via under the each pad so you can reflow them with an iron from the underside if you want to try that.maybe. depends on the PCB layout and the size of the pads. if the pads are like in the spec sheet, it will be tough to heat up the pad and get the solder to flow under it.>>2998991 are you manufacturing the PCBs yourself? If so, just make the "outside" part of pads slightly longer and then you might be able to bring the iron to it.but still, a cheap hotplate or hot air is much easier. hell, even a clothes iron with a PID would do for this.
Sorry for dumb question but is there a technique to locate a short component in a circuit with just a multimeter? I know how to check if there is a short circuit and I'm also aware of the current injection technique. Just wondering if it is possible with just a multimeter.
>>2999075Probe for resistance.https://www.youtube.com/watch?v=GnxO3gwf8E4
>>2999075A multimeter’s resistance mode injects a current, but a pretty low one, and without 4-wire sensing, so it’s not really useful below a few ohms.A spare battery and a current limiting resistor (and a clamping diode) is cheap and pretty effective. Replace that resistor with a constant current circuit (e.g. LM317) if you want something more quantitative.
>>2999075>if it is possible with just a multimeterif you can read millivolts with good accuracy, make the CC source this guy describesbut arm yourself with patience as this asshole is ponderous as fuck, and has an incomprehensible accenthe made a second video to defend the idea that this circuit is NOT dangerous to componentsbut it's a bullshit explanation, as it only applies to probing the power pins, not regular IO pinsso, proceed with cautionhttps://www.youtube.com/watch?v=P_2GGNr4q1s
>>2999112If you clamp the voltage with a diode, the 0.6V won’t be enough to force open any transistors or diodes, so no current should end up flowing through silicon. Even better if you use a Schottky diode. There might be rare cases where internal silicon connections are broken out externally, like the two source pins of some power FETs, or the multiple ground pins of a microcontroller, but I suspect these are not likely to be in the path of your current. My diy current source has a 10mA, 100mA and 1A range, using the 10mA range makes for a convenient preliminary test to not blow anything up. Though honestly I just use my adjustable bench supply instead, despite its massive output capacitance.
How do you measure capacitors, resistors, inductors in-circuit without desoldering them? Is there a good way? I've been using my multimeter to troubleshoot a fault and my readings are all over the place. Definitely not the specified values. I suspect this is because I'm measuring them in-circuit. Desoldering them would seem like a massive PITA.................
>>2999300Try swapping the leads of the multimeter if there's an active component interfering. Suffice it to say your circuit should be off when doing that.Resistors usually have a labeling so you don't need to measure. Just lookup the color code or smd resistor code.Caps are often in parallel so there's not much to do here.If you have to measure the value of an inductor you have bigger problems.
>>2999300With one of these
>>2999309How do these work? I assumed measuring ESR was complicated since the impedance of the capacitor has to be disentangled from the ESR, and that ring testers were easier to make. But that’s just got a moving coil meter so it can’t be digital.
man i wish there was a clamp meter comparison table, like there is that eevblog forum meter comparison tablei want a dc clamp meter with min/max feature to chuck in the toolbox to replace my min/max handheld meter and my shitty dc clamp meter (e.g. uni-t ut203+)dual-screen ones are also nice since you can often see both ac and dc current, or ac voltage and frequency, but they are expensive (e.g. ut219ds)
>>2999309>>2999310Oh I see, if it's for caps over 1uF, 1uF @ 100kHz is 1.6 ohms, so it becomes a rounding error. 1uF and up is useful, though some really small motor caps are more in the 470nF range, so having a mode that goes up to 1MHz might be nice. Seems simple enough to DIY something with a battery, which plugs into your multimeter for it to measure the output volts. If the meter can measure AC volts at 100kHz then it's really easy, but mine doesn't go above 2000 so you'd need a rectifying stage with minimal dropout, a high-speed op-amp ideal rectifier would probably be pretty doable. To get DC volts proportional to ESR, you'd be amplifying, rectifying, and measuring the voltage across the capacitor, and for that to be linear you'd want a constant current AC source feeding it. That should be pretty doable with an op-amp transconductance amp circuit, but if the voltage range is 0-100mV compared to the voltage range of the op-amp (e.g. from a 9V battery) it will approximate a constant current if using a current limiting resistor. Clamp it with a pair of (schottky) diodes. I guess a square wave would be fine, all that matters is the voltage amplitude is pretty constant, though a square wave would run up into the slew-rate and GBWP limits of the rectifier and/or amplifier.Just looked up some schematics, I see one using a 400V rated DC isolating capacitor so it can measure caps while circuits are still live, seems sketchy but it could definitely be useful.
First board and i need a sanity check. I got a DIP28 and the dimensions between the holes are tiny. Is this, pic rel really the right way, or should I rather route three traces around top and two bottom?
>>2999489What i mean by routing around.
>>2999489Check your PCB fab-house's tolerances. It should be fine, but if it's not any inconvenience >>2999490 looks nicer and is more forgiving of problems. Be those fab-house problems (unlikely) or soldering bodges. You've got thicker traces for your power pins, the only issue I see is those pads making big cutouts in your ground plane. That might be fine, but you could consider reducing the ground pour clearance value so it fills between holes if EMI is a particular concern.
>>2999489>Is this, pic rel really the right way,yeah, DIPs have so much spacing i've even seen 2 traces go between pinsbut i'd space out the 45-deg sections so they're not hugging the holes so much
>>2996880anyone know good resources for thinking about and applying laplace transforms to circuit and amplifier design?i know fourier theory in all its different forms well, however i don't have a good resource for laplace transforms.they seem based more on analysis than linear algebra (even though they are linear transforms) and i don't know how to practically use them for engineering.
I want to convert some old organ pedals into continuous midi controllers.It's pretty simple in terms of construction, a hole where a switch used to sit is pushed into by the pedal, and I'm trying to think of the best way to measure that gap.At its lowest, the pedal sits about 11mm low. At highest, it goes maybe 4mm below the hole, and it moves in a slight arc, so about 7mm of travel. I initially thought hall effect would be good for this, but it seems maybe slightly far to get a good reading, and they seem like a pain to work with. I'm now thinking some kind of optical sensor would be better, the holes are fairly closed off from each other.What's good in this field? I want to spend about 1-3 bucks per sensor, and ideally use something that runs off 3.3v that just has an analog output.
>>2999680Guitar wah pedals move a potentiometer through mechanical linkage (gears).
r8 my breakout matryoshka, /ohm/>300x400 mil protoboard with pin headers>300x200 mil spacer/insulation board>TLR377GYZ-EVK-001 breakout board on top of that>TLR377GYZ on the very topthe chip itself is the world's smallest opamp. 0.88x0.58 mm 2x3 0.3 mm pitch BGA
>>2999688B-Higher grade if you remove the breadboard from the equation.
>>2999685Yeah, but potentiometers fail, and I want something that can deal with very fast movement. I saw a paper where a guy used VCNT2025X01 for harpsichord key sensing, so I bought some TCRT5000 boards because they were 30 cents each.
>>2999688>world's smallest opampthem giantastical wires hardly match the theme of the partytry 30AWG wire wrap wiresa wire so thin, it's almost wirelessshould always keep some at ohm
>>2999690You could use a rotary encoder, but it's digital. FWIW, a wah pedal can last a long time if it isn't abused and cleaning a pot isn't too much of a hassle.
>>2999698The main issue here is the hassle of gearing up 20+ pedals. They all pivot on one very long shaft.
>>2999696I do have some really thin wire but stripping it is more of a pain than using the 1.27 mm stuff
>>2999700IR emitter/sensor combos could work. It might be easier to mount them at the lever where the motion is more linear and the feedback is smoother.
>>2999680>some kind of optical sensor would be betterif the pedals are thin enough, an opto interrupter gives you a much cleaner and stronger signal than the reflective stuffand it works fine on dark surfacesno discrimination up in this bitch
>>2999720Not really possible, the hole at the top is smaller than the pedal width.
>>2999680If you’ll be interpreting it with a microcontroller anyhow, use a static hall sensor and have alternating poles of magnets on a rod that moves past it. If it’s a binary hall sensor, then you get a low resolution. If it’s an analog hall sensor, you’ll get plenty of resolution if your ADC can handle it. If you use a digital rotation hall sensor, like in the Open Micro Manipulator:https://github.com/0x23/MicroManipulatorStepperYou can achieve sub-micron precision.Think of it like a non-contact rack and pinion, where the rack teeth are the alternating magnets, and the pinion is the sensor.Equivalently, you could make other custom encoders. Say you have two or more static photointerrupters, and a series of cutouts move past them.>>2999721Gonna have to post a picture or diagram or two of the kind of space we’re working with.
>>2999724Seems like too much effort when I'm crushing it all down to 127 points anyway due to the MIDI spec. If I can get 0.1mm resolution over 7mm out of these cheap IR combo boards, that's probably good enough with a bit of software smoothing.
>>2999720I got tons of these from an old printer someone threw out. I once got a free printer when I bought something at this computer store, but the ink refill cost $100 so most people just throw out their inkjet printers and you can find parts in them. Newer printers, less-so though.
latest Veritasium is about the Smith chart: https://www.youtube.com/watch?v=GK2pZ_oVU1o
>>2999726>If I can get 0.1mm resolution over 7mm out of these cheap IR combo boardsWhat sort of combo board are we talking about?
>2200w sewing machine ac servo motor i just got this chinese motor and id like to put it on a drill press with a pair of simple spst foot switches for fwd/rev so the motor would only be on when the corresponding direction pedal was depressed does anybody have an idea if i could tap into anywhere on the controller do accomplish this? im used to vfds that have simple terminals for this basic control and im mildly retarded
>>2999767heres the existing control interface
whats the trick? what kind of a flux do you need to solder batteries?
>>2999771probably zinc acid like plumbers use
>>2999767I assume the D button is for direction, that will be a pain to interface with via pedals, so I'd want to bypass it.Do you want variable speed from analog pedals? Or just on-off pedals with speed adjusted by hand? The former is difficult. The latter just requires proper wiring of the foot switches. For a brushed DC motor you'd just need a pair of SPDT switches, but for an induction motor with a run cap, I suspect you may need DPDT switches for each pedal. Or relays if you'd rather not have the full motor current going to and from the foot switches.Assuming the potentiometer knob on that controller is for speed, If you do want variable speed analog pedals, you'd need to put a potentiometer in each pedal using a rack+pinion or pulley mechanism. In addition to the switch. If instead it's digitally controlled with the +/- buttons, you'd have to get in there with a microcontroller and some transistors or optos or reed relays to fake the button presses. And probably some feedback measuring the display segments to figure out what speed it's set to. Not easy or fun at all, to the point I'd be trying to replace the controller with something more hackable.If you can elaborate what sort of motor you have, and how the controller's interface works.
>>2999788i wish it was just cutting the leads to the control but thats a little board itself with a logic chip and the outputs arent directly connected to the buttons
>>2999788>elaborate what sort of motor you haveits a generic china ac brushless servo there were originally sold for sewing machines
>>2999793I mean, just take that board out and run some wires to where the button contact is? The easiest option would be just to have one pedal hooked up to the potentiometer for speed control, and a second foot switch to swap direction. Although I don't know what that pot actually does, you might have to go really simple and use 3 foot switches, one for speed up, one for speed down, and one for reverse direction.
>>2999796thers way more going on with the little controlboard than on/off since it interfaces the parameter settings too. the direction control button isnt instantaneous either. it goes through a ramp down then stop then accelerate sequence and clicking the button during the process does nothing. and if the speed pot is set to 0 the direction control button doesnt change the starting direction. its also used to confirm a parameter change
>>2999611anyone?
>>2999797Yeah, I'd just interface them all out to ports and hook up some labelled foot pedals.
>>2999793>>2999794Ah, so it's not an induction motor but a proper synchronous motor with a variable frequency drive, with its own encoder. You might be able to swap phases AND swap encoder wires in order to swap directions with mechanical switches and no extra logic, but it depends on how the encoder is wired.Again, my questions stand:>Do you want variable speed from analog pedals? Or just on-off pedals with speed adjusted by hand?>How does the controller's interface work, especially the knob on the front, is it a potentiometer or a rotary encoder?
>>2999611Isn't that control theory? Poles and zeroes, stability of feedback loops, that sort of thing? I haven't studied it myself, but there seem to be some freely available MIT lectures on the topic. At least I assume control theory uses laplace transforms. Maybe digital signal processing does too.
>>2999771don't solder batteries, dumbassif you must, use a low-melting solder like indium-tin>>2999611laplace is just fourier with a real term added. it's actually shocking that we don't just teach students the laplace transform immediately seeing how useful italso look up Brian Douglas' control theory lectures on youtube
>>2999771spot welder and nickel stripsaccept no substitutes
>>2999611>>2999798books on analog and digital filters.
>>2999771>whats the trick?the trick is to use a file or tiny grinding wheelthe metal has a zinc coating which will not wet solderyou file off a small section near a corner to reveal the steel beneath, and solder to thatsteel is not great at receiving solder, unlike copper or brass, but it will work eventuallygive the wire a tug to make sure it's solidly James Bond'ed
>>2999839There's actually a cheap jig you can buy specifically for that.But really >>2999813 is the way to go.
>>2999774>probably zinc acid like plumbers useI'll try to find some of that.>>2999812>if you must, use a low-melting solder like indium-tinwtf is that shit so expensive? I can buy a spot welder for that.>>2999813>accept no substitutesI don't want to wait 2+ weeks for it to arrive from China.>>2999839>the trick is to use a file or tiny grinding wheel>the metal has a zinc coating which will not wet solderThanks! I'll give that a try. I have find sand paper.I need to solder it today.
>>2999771For low current applications just use a 18650 holder
>>2997449update on the pee hole circuit. I've ordered boards and parts
>>3000125Mind telling us what this is for? Soviet spy microphone?
>>2999489>>2999490the parasitics in this circut must make tape worms jealousim guessing this is a 5V circut with no more than 10mA per pin and bus speeds lower than 20MHz so must be negligible
Was designing a CC/CV feedback loop for a 4-quadrant power supply, operating off 12V like a deep cycle battery. PWMing an H bridge with a pair of inductors and capacitors to make a bipolar buck converter from a single supply, using a conventional triangle wave comparator PWM circuit. Turns out LM324 differential amplifiers work for everything. One as a current sense amplifier with a gain of 10, one as an attenuator for the output voltage with a gain of 0.1, both referenced to 1V, as the DAC outputs go from 0V to 2V. With error amplifiers with a gain of 100 comparing those measurements to the DACs.But I just realised that my method of combining the two feedback loops with diodes won't work. Normally, if your DACs are set to +5V and +0.01A into a 100 ohm load, the 0.01A output would pull via a diode to stop the output going any higher, the voltage would stop at 1V, and the voltage feedback loop would be open-loop. But then if I tune the current past 0A down to -0.1A, the current would go up to -0.1A and the voltage would shoot down to -10V.What would be the best way to solve this? I know I could use the MCU to drive the H-bridge directly, but I'd rather have analog feedback loops. I could measure current and possibly voltage before the H-bridge, but then I'm losing information, which might matter especially when I'm doing some sorta galvanic cell or whatever that operates in the generating quadrants. There's a few power op-amps (e.g. LT1970) with integrated current limit that people are using as 4-quadrant linear supplies, in the same vein you don't appear to have any ability to specify whether the set current is negative or positive.More op-amps? MORE OP AMPS???
>>3000126converting capacitance to frequency
>>3000163you could build a max() circuit maybe. that way regulation happens on whichever of current and voltage that is "too high"
I've used KiCad 7 a little. I just loaded up KiCad 10 and plan to make a few projects. I don't feel like I need a "Drawing Sheet" to build my schematic. Any thoughts on that?
I feel retarded here installing a vent fan with a built in humidity sensor. I have 2 switches in the wall and don’t want to run another circuit. I’m going to put my light and fan on the same hotleg so they just come on together. I have no idea what to attach the brown and black wire to though. The red wire coming in is from the other unused wall switch so I know one of the brown or black wires go to it, I just don’t know which one and what to do with the other wire.
>>3000188What are you hoping the second switch will control?What is the unwired black wire labeled? What is it going to?The brown is labeled SC/H/M, right? It's hard to tell. What is the brown wire going to?
>>3000188Also, do you have either a schematic or a part number for the fan/whatever the brown and black wires are going to?
>>3000188It would help if you had a make/model, but I'm guessing that the humidity sensor (brown wire) is supposed to be unswitched because it controls a relay to the fan which is the black wire (hot to fan). So you either switch it at the wall, or the sensor switches the relay when humidity is high enough.Brown - unswitched hot supply for the sensor and relayBlack - relay-switched hot supply to the fan*It might be vice-versa. Read the instructions.
>>3000192>when humidity is high enough.and/or shut off when the humidity decreases
>>3000189I want to keep the second switch on at all times and it’ll just kick the fan on when the room fills with humidity. The brown, 1 black, and 1 neutral are bundled with the SC/H/M label. >>3000189>>3000190It’s a Kaze SNP100HLP. I’m not an electrical guy, but it looks like the brown goes to the red hotleg coming in. But that leaves me guessing what to do with the unused black. Does it go to the other black hotleg coming in/motor/light? Or do I just cap it.
>>3000188>I’m going to put my light and fan on the same hotleg>>3000194>I want to keep the second switch on at all times and it’ll just kick the fanI'm confused; these two statements seem contradictory. Can't you just wire it as your schematic shows?
>>3000194Tie all the black wires together. Tie the brown to red and keep the other wall switch on so the sensor works, or bypass the wall switch and hard wire the sensor in a junction box.
>>3000194I think you need to tie the black Motor wire to the black SC/H/M wire. So wire as the schematic shows omitting the fan switch
>>3000195I want the fan to come on when someone is pooping with the light on. But I also need it to come on when someone showers and the room fills with humidity. It’s a bathroom toilet closet inside the master bathroom. It’s gets very damp in there when showering and I need the fan to come on when we forget to flip the switch. Schematics show 3 switches in the box and I only have 2. >>3000196Ok. That’s the way I was leaning but didn’t want to fry the motor. I’ll give it a shot and see.
>>3000198Brown to red. All blacks and light togetherAll neutrals together. https://youtu.be/dHTBPa5fDkk?t=116&si=7rOQ06xu86cWgRfg
>>3000198SoIf the light is on the fan is on.If the light is off and there is humidity the fan is on.If the light is off and there is no humidity the fan is off.Is that what you want? If so, you need a 120VAC coil relay where the coil is controlled by the "LIGHT SWITCH" and the contact is the "FAN SWITCH" above "COM". Everything else would be the same as the schematic.Or you can do it without the relay by tying the black wire from the light switch to the blue of the light and the black of the fan. But then your light will turn on whenever there is humidity.
>>3000199If you do this your light will turn on when there is humidity
>>3000200Yea pretty much what I’m looking for. I may try the relay later if this >>3000201 gets annoying. Thanks everyone for the help. Really appreciate it.
>>3000202Anon, do the following:1. Tie all blacks together with blue.2. Tie red to brown.3. At the wall switch, jumper red to black or remove the switch and install a junction box so you can permanently wire the sensor.Then you can poop with the fan + light or shower in the dark and let the sensor switch the fan.
>>3000206or leave the switch and put tape on it so nobody turns it off.
>>3000207The humidity sensor controls a motor relay (built in).
>>3000209Yes, but he has two inputs for one fan.Humidity->fan onLight switch->fan onThose are not the same input
>>3000210One hot is switched at the wall. The other is switched by the sensor/relay. They are independently switched. Otherwise, the sensor has to be switched on which is confusing.
>>3000212>Otherwise, the sensor has to be switched onThis is a good thing because then you can turn the sensor off if you need to. Like if you need to dust the fan blades or get a bird out
>>3000214You can do that when the humidity drops and the fan is off. lol
>>3000215True
>>3000172A min() circuit is what the normal diode combination of error amplifiers of a CC/CV supply normally does. And as I said, -10V isn’t as high as a +5V limit.I should probably draw out a chart of all the possibilities for set-points of voltage and current in the four quadrants, and different types of loads. Because it’s for electrochemistry, the load could be positive voltage but negative current.>>3000183What you do is ignore the writing on the sheet and build your schematic anyhow. There’s probably an option somewhere to delete or at least hide it.
>>3000163>>3000228I guess the question I have to ask myself is the following:>current setpoint = 0.1A>voltage setpoint = -5V>10 ohm loadShould the output be -5V, -0.5A, or should it be 1V, 0.1A?Equivalently, what about when the current setpoint is 1A? Should the output be -5V, 0.5A, or should it be 10V 1A? The same questions apply to a negative load resistance, but for test values in the positive quadrants.I think I’d prefer 1V, 0.1A for the first, and -0.5A -5V for the second, but that’s kinda arbitrary, I don’t want it flipping quadrants just because the effective resistance passes through some non-zero threshold. So I guess the only answer is to ditch the min or max diode logic, and instead digitally control whether it’s operating constant-current or constant-voltage. With an analog switch IC, JFETs, or just something simple like transistors pulling the signal down. I could still pipe the sense amplifier outputs into ADCs to measure voltage and current if I want to add software limits.
>>3000243Actually I probably could have analog switches with multiple diode combinations so it can smoothly switch from CC to CV, or whatever. At least I think that makes sense.
Dumb question. How hard is it to get 6 volts from USB-C?I have an old device (Key Digital KD-CSW4x1 component video switcher, picture from Google) that is missing the original power supply. For some odd reason the OG PSU was 6 volts at 5 amps. That seems like a lot of electrons for a switcher but it apparently was a very good one for its day so whatever. A new chinesium PSU with the connector I need is $15. Assuming I can source a connector, can I cheaply get 6 volts out of USB-C or would that be not worth the cost?
>>3000358>How hard is it to get 6 volts from USB-C?6V is not a standard voltage. Arguably you can use USB-PD 3.0+PPS to get 6V at 5.0A, so long as you have a PD supply at that supports PPS at 5A, and a 5A rated PD cable, and a PPS-based PD trigger module. That's the hard part, I haven't seen any off the shelf PPS trigger/decoy boards that aren't kinda expensive, bulky, and with a screen and buttons. No clue if they even have the ability to save their programmed state. PPS is also kinda strange, you can't tell it "give me 6V" instead you activate PPS and tell it to increase or decrease the output step by step, so arguably it's not a very robust protocol. Plus it will be hard to find a 30-45W PPS-capable charger. Even if you can get it to run off 5V directly, a lot of 30-45W PD supplies are 3A maximum. It is still possible, with a custom PCB you could probably get a PPS trigger down to $5-12 per board, in lots of 5 boards, using one of those chinesium PD trigger capable MCUs, if you're willing to do that programming. If you need more than one of them, or want to sell them, or have other uses for PPS PD, go for it.Instead I'd probably lean towards a 3A@12V/15V PD trigger, followed by a buck converter, if not just buying a replacement PSU.I'd also look inside this machine, specifically why it has a 5-pin DIN plug. Maybe it's getting split-rail +/-6V, or +/-3V. A dual supply like that would be useful both for RS232 and analogue signal handling.
>>3000358I would just try a 5VDC PSU. Maybe it's 6V to compensate for voltage drop across a reverse-polarity protection diode.
>>3000358>>3000367just get a small USB-C power delivery board and set it to 5V and then boost it to 6V with a trimmer pot. pic related is an open source "PPSTrigger - USB PD Trigger SuperCharged with PPS" but you're better off getting one pre-made from China.
>>3000369yeah, this might work too. but I've had some issues in the past where 5V at the input was not enough to trigger some TTL ICs but 6 or 7V would. maybe he could remove input protection diode in such case and 5V would work just fine. no idea how far the anon wants to go.if not, something like >>3000370 is the way to go.
>>3000370Where do you get those from china? I'm kinda surprised I haven't seen any clones of those on alibay, it's been at least a year since that design was published.
>>3000371>remove input protection diodeAnon could inject 5V after the diode to test. If it works then just jumper the diode.
>>3000372>Where do you get those from china? I'm kinda surprised I haven't seen any clones of those on alibay, it's been at least a year since that design was published.the author of that wants <$10 on Lectronz.on Chinese sites look for "MT3608 Boost" boards and find one with the Type-C plug. It has a trimmer.>>3000373>Anon could inject 5V after the diode to test. If it works then just jumper the diode.Yep. He should def do that and that's the way to go (if the diode is the cause for the 6V req(. the USB-C board will act as a (much better) protection anyway.
>>3000375>MT3608 BoostNot at all the same, it's just a 5V non-PD board with a noisy chinky boost convert on it, and it's only rated for 2A. A PD supply wouldn't even provide 5A to something without a proper PD identifying IC on it, and for 6V@5A you'd need at least 5V@6A, which isn't USB PD compliant at all. You'd be better of stepping down from a higher voltage.
>>3000367>specifically why it has a 5-pin DIN plugThat is actually a picture from a different revision of the device. Mine has a "standard" barrel connector, as much as there can be. There are a couple dozen "standard" barrel connectors out there in common use.As for your suggested solutions, way outside my wheelhouse. I'd hoped it would be something simple I could bodge together using a cheap USB charger and a hacked together cable. Kind of like how you can power some old Atari computers the same way.
>>3000381>>3000367>>3000369>Maybe it's 6V to compensate for voltage drop across a reverse-polarity protection diode.I said, fuck it, it will probably have a chip in it that handles power. Open it up, find the chip, and google it. Twenty FUCKING three screws later, I have the inside of the device. More to follow.
>>3000382I am not really an electronics guy. I can solder well and have repaired a few old computers/consoles by finding bad chips and replacing them. When it comes to doing actual electronics stuff, not just fixing what is there, I am a dunce.That said, I can google IC numbers! It appears to be a LM2940C. TI says this thing takes input voltages of 6 V to 26 V. So, I could probably use a dirt common 9v or 12v PSU! Yay. I hope they didn't go with the minimum voltage so they wouldn't have to put a heatsink on it or a fan in the case.
>>3000383No traces on the bottom of the board. Multilayer PCB. Huh. Didn't expect that.
>>3000384The other end of the top side of the board, in case anyone wanted it.
>>3000382>>3000383You could use an external buck converter to bring 12V down to 6V so the regulator doesn't get too hot.
>>3000382>>3000383LM2940 has a minimum input voltage of 6V, but it's a 5V regulator so you should be able to bypass it if you run it at 5V. But it's only 1A rated, so it can't be where all the current goes. Overvolting its input would make it dissipate more power as heat, since it's a linear regulator.The real question is, what if anything runs at the native input voltage.>>3000384You can still look up other IC numbers, see what their voltage requirements are, and measure continuity from the +6V input to their Vcc pins.>>3000386Yeah, this or use a cheap PD trigger set to 12V.
I'm repairing my 35 year old stereo amplifier. I have a few of questions for you guys:1. The original speaker protection relays have been discontinued and are no longer available. Do you reckon if I remove one of them and go to an electronics store they would be able to find a matching replacement?2. The two 6800uF filter capacitors tested out fine for capacitance and ESR, but I think they look very slightly bulged. Should I replace them?
>>3000431Also, I heard somewhere that all fuse resistors should be replaced. But I think they were talking about Sansui and I have a Technics amplifier. Opinions on this?
>>3000431do the relays have any ratings on them? if so get replacements that meet those ratings. current is likely to be the most important thing
>>3000431Relay replacements should be doable. Absolute worst case scenario, you have to buy relays online with a different footprint and missing an internal component or two (e.g. snubber network or diode), using wires to donned them to the original relay’s pads. But relay footprints are surprisingly standard, so I suspect you’ll be fine. Local stores vary a lot, often they just have automotive-style relays. Do a search yourself before desoldering them, filter by maximum voltage and maximum current, to be equal to or greater than the old relay (or what the circuit will actually see), and also probably by dimensions, number of leads, etc. The parametric search tools on sites like Mouser and Digi-Key are great.Even if the caps are measuring good now, with a bias near 0V, they may not have low leakage at higher bias voltages. Unless you can test their leakage current at higher voltages and higher temperatures, I’d want to replace them.>>3000432No clue about fusible resistors.
>>3000387Eh, that is a lot of work to save $15. That you for the help though. Also, literally every IC except the power IC has had the numbers scratched off of it. Haven't seen that since I replaced the RAM on some old 80's arcade games. I guess they didn't want anyone to copy the design. I found a manual for this thing while searching for a schematic (which isn't available, natch) and this thing supports composite, S-Video, component, and RGBHV. Crazy.
>>3000431>remove one of them .. electronics store ... find a matching replacement?if you mean an actual store, the chances are very slimonline, you get a much wider selectionbut none of that mattersall you need is match the coil voltage and current capacitythen solder 3-inch long wires to each terminalturn relay upside down and hot glue him/her to the PCBas for the caps, leave 'em alone unless you hear hum at high volume, with a shorted inputreputable companies will over-rate their caps so they'll work ok even as capacity drops with age
>nordic released a new MCU with a built in USB-HS PHY for like $5>keychron just out an empty repository for their own Zephyr based open source mouse firmwarefuck i guess it's time to restart that diy mouse project i gave up on
I bought a lot of 20 broken laptops. They're all the same model of a Thinkpad. I need to fix them to earn back some money from this investment. I've been looking to get a thermal camera to make my life easier.What resolution or how many pixels should I be looking out for? I know fuck-all about thermal cameras.
>>3000625Cool. How will your design differ from Ploopy's mouse designs?>>3000631Cheap&Cheerful on YouTube has a series of review videos on thermal cameras. For electronics troubleshooting I'd be looking at one with a macro lens. The USB C models are definitely a bit cheaper than the standalone models.
>>3000631A common consumer size is 256x192 pixels, it works well enough.I'd highly suggest going for a Chinese one. Cameras by companies with a US presence are capped to 9 FPS by law, meanwhile Chinese ones are often 24 FPS. Cheaper too. Some USB-C ones present as a webcam (be sure to check), so for those ones you can use open source software if the app sketches you out or is ever discontinued (I only run mine in airplane mode, and backed up the apk).Mine is an Infiray P2 (pro version has a macro lens). While mine works great, dunno if that's still recommended. Aside from the usual shitty marketing and astroturfing chinese companies do, I saw apparently a newer revision wasn't as good. Either way, most of the cameras with those specs use the exact same sensor so pick your poison.
>>3000631>>3000633>>3000635As well as the P2 Pro, it looks like the Thermal Master P3 and Topdon TS001are also pretty good value dongles for macro work.https://cheerful.cheap/category-results/
>>3000633>How will your design differ from Ploopy's mouse designs?wireless, USB-HS (8kHz polling), decent hot-swappable switches, at the bare minimum. ploopy is a terrible design that was obsolete the day it was released
>>3000641>decent hot-swappable switchesA total meme. They are retarded for using the cheapest omron in the range instead of just putting the 20M version in, though. 5M switches die in 3 years, 20M last at least a decade.
>replace three capacitors on gen2 prius combo meter>change xbox xclamp and thermal paste>replace 2010s iMac storage and upgrade RAM>replace rotted plywood floor in rapeshed>route CAT cables for camera system I wish I had friends that were interested in this stuff
>>3000662Or just use magnetic or optical switches that last indefinitely.>>3000665I wish I had friends.
>>3000631I bought a cat s61 on ebay, its nice having a full relatively ish recent android device behind the flir sensor for moving the images to my computer or w/e rather than a chinky clunky proprietary standalone ir cams OS or shitty plug in thermal cam.
>>3000678The plug-in cams can be used with a variety of 3rd party apps as far as I understand, and they can also be used with a USB C extension cable to connect to a computer or tablet or whatever. Less convenient in form-factor, but definitely more flexible. Also when you upgrade your phone because it stops getting updates in under 5 years, or the battery stops charging properly and replacement parts are hard to find, you don’t get to transfer the thermal camera to a new phone.How does the Cat phone compare to the Ulefone thermal camera phones? I like the look of the mini 20t pro, but apparently their ROMs are kinda locked down.
Where do I start building my own high quality microphone? I don't need it to be aesthetically pleasing. Just need to look for reasons to sniff my leaded solder fumes.
>>3000691Depends on what you intend to record. Different mics for different situations. Dynamic, condenser, ribbon.
>>3000690logical fallacy, you and i both know these two options can become unusable with time.However I would HIGHLY recommend against these cheap plug in cameras since the apps are always locked to the camera and the apps become unsupported, adware bloated, broken by android/ios updates, etc. Having a dedicated device at least means it can be locked down to never get updates (e.g., turn on airplane mode) and thus never break software wise, the typical battery etc issues come with a standalone device and thus is invalid in argument.And lol no chink phones i tried with thermal came even remotely close to being as well packaged as the cat s61, even though low resolution, it has a beautifully calibrated accurate sensor, these chinese ones have +/-5 degF variance at MINIMUM, plus poor support etc and LOVE to randomly die/lose alignment.
>>3000691>Where do I start building my own high quality microphone?consider this monstrosityhttps://www.youtube.com/watch?v=LoQu3XXIayc
>>3000633>For electronics troubleshooting I'd be looking at one with a macro lens. The USB C models are definitely a bit cheaper than the standalone models.ty!>>3000635>A common consumer size is 256x192 pixels, it works well enough.>capped to 9 FPS by lawthat's crazy! they're handing over the market to the chinese>>3000640>Thermal Master P3 and Topdon TS001areThanks! P3 was already on my radar. I'll see if I can get a deal on it.>>3000690I setup an alert for it. If I comes up for sale locally, I might get it. But I'd rather buy a brand new USB-C camera since it's cheaper and probably better.
>>3000740dont do this it will sound like ass in an untreated room without a proper capsule
Guyd, look this circuit that I'm breadboarding. It's a two stage headphone guitar amp. The first stage is a buffer done with a BF245C jfet and I biased it with a voltage divider for more room for the signal to swing without clipping. My jfet has -0,88 fot the Vp and 20 mA for IDSS. I managed to amplify 10 V without distortion. Vcc is 12 V. The second stage is a current buffer done with a BC548B bjt transistor. It is running hot! There's too much voltage over the colector and emitter, 10,3 V. I biased arround 25 mA at the emitter because this gives me around 3 mW for my 11 ohm load. It had to be two stages because the jfet was loading the headphone. It's class A! I wanted to be class A for the novelty of being class A. It's not perfect, but it's the best one I did so far. Until now I could not get a crystal clean sound. The circuit always had some brekp up in the sound, like a distorted valve, which is nice, which is fucking cool, however, I wanted to have a clean signal. This is the best one so far. I just added a volume pot to kill some of my guitar signal, and before the pot I put a resistor to limit even further. It has room for improvement, but I did guys. I designed all by myself. I did all the math. I did not copy anyone. I mean, I had to use the same building blocks that we all use, but it's my design. It's orignal. I had to measure the components and shit. It's so fucking frustrating. My electronic course lied to me. They lied to me! The whole system is a lie! Our educations sucks! They just want us to know the basic stuff.
>>3000845congrats on the finely engineered circuit working as intended. there is no substitute for actually sitting down and building the thing
>>3000845Here's the diagram. The real life circuit had to be different because the JFET in this simulation did not have the same parameters which my real life jfet has so I had to measure the parameters in the simulation and design the simulation with those parameters and in my real life circuit I had to do the math all over again to change for my real jfet. I was thinking about chaning my buffer. If you see the diagram I had to kill the signal with a resistor, this is not ideal. The signal is comming out of the buffer as it is, very close to 1/1. But my current buffer is design for small signals because it is for headphone. If a big sinal comes in it will want to have a high current to drive the small 11 ohm load. I was thingking about killing the signal at the buffer. I will use the math at my advantage. If I lower the gm, transconductance of something, I can make the buffer atenuate the signal. Let's say 2Vpp comes from my guitar, if I do the math correctly, then the JFET will kill all that signal and atenuate it to the proper small level and I can remove the resistor after the output of the first buffer. This is why I said it can be imrpoved. I can also add a current source at the current buffer. But I wanted to be simple. Just two transistors. I think it's simple, sort of. The BJT is running hot. But I don't care, I want to see how much this transitor can last with this amount of heat upon it. Let it suffer. Let's see how much heat it can take before breaking up and dying. Just like life does to us. Gos is testing us right now. The fucker. Fuck him. He won't breake me!
>>3000846The pickups response for my breadboarded amp was surprisingly good! I have a guitar with 5 positions and I was amazed that it preserved the tone of all the positions! Before, with the others ciruits, it was not like that! There was no response. I mean, you could sort change the positions and the knobs and kind perceive a little change in the sound, but with this new design it took to a next level. The buffer gave it the power to perceive even the pick! The sound of the pick is amazing. You can play very percurssive with this buffer. It preserves the quality of the picking, so good! I try to be inspired by the Fender Champ amplifier? It has only one volume control, right? So I tried to be like that in this sense, to have at least one volume control which acts as a breaker to the circuit. As you increse it breaks the circuit a little, I think it adds a little bit of compression, but this comes from the BJT transistor. The JFET buffer does not distort. Anyway, I was playing some blues licks. It is good to play the blues when the volume is totally full and when you lower the volume it cleans and it's cool as well. I think I managed to make it more clean than ever before. Even with the volume full up it has some distortion, but it is still clean. It's not heavy distortion and I can still play bar chords which sound clean, but a little bit bigger and fuller and rounder? This is why I think at the full volume it is distorting a bit. But, I'm happy. My dick was hard when I first listened to it.
>>3000847>The real life circuit had to be different because the JFET in this simulation did not have the same parameters which my real life jfet hasthis is a common problem with JFETs. the pinchoff voltage has a huge range, so for certain circuits you need to measure and bin them>>3000848one fun thing you can do is add a pot to deliberately distort the output. like changing the bias on the JFET stage. this follows a typical "workflow":>build a thing, don't know what it's doing but it sounds cool because of distortion>build a thing properly, not it sounds clean>build something like the first thing again, but now you know *why* it has distortionI've gone through something similar with my pee-hole Wien bridge oscillatoralso catbox some of your guitar playing pls
>>3000847Damn that’s some high input impedance, I’d expect excellent reproduction of sound from such a high input impedance, without loading down the pickups or tone circuitry, but it will make it more prone to noise. You could try it with 820k/330k to see if it changes the sound.Is an emitter-follower called a class-A amp? I’d usually not bother with a DC-blocking cap before an emitter-follower it I can avoid it, though it was probably necessary to get RV1 working. Maybe you could have got away with just one cap (perhaps from RV1 to ground), but if it works I have no complaints.Also C4 is backwards, its right leg is more positive than its left.
the pee hole boards arrived. they're pretty good, except they plated the area between the tines. will have to file that off when assembling. see picrel. old boards are on the left and are less dense and twice as thick, and use an inferior circuitstill waiting for parts
>>3001008the undesired bridge is marked in red on the second board from the top. the coax is RG-178
>>3001008>>3001009You could put a IPEX/U.FL footprint there, and solder pads on the other side.
>>3001011nope, they're too big. I checked
Imagine a grid tie solar system that’s got old micro inverters that all work but don’t wake up in the morning without a 240VAC power cycle. Power cycling solar is annoying tho so would there be anything wrong with using a baby solar panel to turn on a pair of solid state relays when the sun comes up? As long as they fully open/close with low light conditions I’ll be satisfied. Wouldn’t want partial AC voltage or some not fully on/off state.
>>3001017>using a baby solar panel to turn on a pair of solid state relays when the sun comes up?kind of a bad ideacoz, you'll get chatter from clouds, and false activations from head-lightsyou can get a Kasa smart plug for just 2-for-$15some smart switches even let you choose sunrise/sunset as activation times, which adjusts itself throughout the yearit uses internet time, not light sensing
>>3001107I’m hoping PV voltages will be pretty steady when theyre lit up. the current mostly fluctuates with clouds. I looked at the datasheet for the relay and it says min.2.4V for on and >1V off and 7.5mA trigger. Stole these cells off a light they’ve got to be 50mA at least and 3S 2P ought to do something. Otherwise there’s million timer solutions I could get out of a box.
>>3001125>I’m hoping PV voltages will be pretty steadygood ideadesigning hope into my circuits almost always pans outin my dreams, if not in real life
>>3001125PV open circuit voltage is quite flat over a wide range of illumination. this may or may not work depending on your latitude
>>2996880Trying to learn analog electronics and circuits I have gone through most of the stuff already but I am wondering how can I learn all the new stuff or how modern circuits are designed? I feel like everything I read is from the 70s or tutorial level. Is there something like a library of circuits similar to how there are software libraries and you can reuse them?
>>3001158>library of circuitsYes, and a lot of them are in the form of integrated circuit chips to reduce the number of components on a PCB. Look at datasheets to find the block diagram of the component. It will look similar to pic related.
If I don't care at all about a slight voltage drop (because it's feeding opamp power rails with plenty of room to spare), can I switch an LC filter for an RC filter? In simulations, on the frequencies I care about the response looks as good as, or even better, than with an inductor, and inductors are fking annoying with their resonances etc.
So this AC stopped working after somehow the top tank overflowed and water poured all over the place.When I plug it in the control board freaks out with all lights flashing and beeping and the voltage tension fluctuating between 0 and 5V.The float and temperature sensors seem ok.Any idea of what could be wrong?>inb4 the fuse is ok
Does this shit expire? Works fine at 750 and it was free. Made before I was born unless it's a date code .
>>3001177Oh wait it's 06 not 86. Still almost old enough to go die for Israel.
>>3001176What are those black lines near the relays, IC and coil(s)?
>>3001192that's air gapping to mitigate high voltage arcs i think. they're milled out of the PCB
>>3001194it's to minimize creepage, not arcing
>>3001158Datasheets and application notes for modern parts will tell you how to use those parts. Once you understand the eternal fundamentals (transistor topologies, op-amp topologies, digital logic, power supply topologies, etc.) the tricky part is more Im finding the components that suit your purpose. You’re not going to make a switching regulator out of op-amps and comparators and gate drivers, you’re going to find an IC that implements a known working circuit in a way that suits your application. Browsing a manufacturer’s website (I usually start with Texas Instruments) for their search tools is a better way of understanding what kinds of components are out there, and what kinds of keywords you need to look for even if this company doesn’t make your desired part.>>3001176Maybe a power surge killed the PSU or logic. If the 5V line is flashing on and off, the switching converter could be going into hiccup protection mode, suggesting a short circuit down the line, or maybe a dodgy capacitor. If there’s a power-good pin on the IC, that could help. Measure resistance from the 5V rail to the 0V rail after removing power to the board.I see there’s also a 12V rail on the board, what’s it getting up to? Troubleshoot it too, 12V lines are more likely to be fault-causing than 5V lines I’d guess.Unplugging connectors to see if anything off-board is causing it to fail is also probably worth attempting.That NTC thermistor is also possibly suspect, make sure it’s measuring as it should.
>>3001008You didn't get them to assemble it for you? Gonna use an oven, hot plate, or hot air?>>3001163Absolutely. In fact, they're often recommended because they don't cause voltage overshooting like LC filters do. I often see them with resistances like 100 ohms. A third option is to use an inductor and a resistor, and deliberately make the RLC circuit low-Q, to prevent ringing. That's more important if you care about high frequencies, ferrite beads even more so.If you need a bit more power output or a lower corner frequency without a fuckhueg capacitor, I'd go for a capacitance multiplier. Really just an RC filter with an emitter follower, the 0.7V dropout is typically not a problem. Conventional linear regulators are also worth considering, but you'll likely get more dropout, and their ripple rejection outside of 100-120Hz isn't brilliant, but the have a well regulated voltage which you may appreciate. I calculated that a 7812's ripple rejection, combined with my op-amps own PSRR figure, was enough to trivialise my power supply noise even though it was in the 10s of kHz range.Also big regulator doesn't want you to know this, but JFETs work as linear regulators. Their gate threshold voltage intrinsically acts as a voltage reference for what becomes a source follower. Often in the handy ranges like 3-9V. They probably drift badly with temperature, which means any FET low dropout is moot, but I suspect their ripple rejection is pretty good.
>finally solder up project after days of double-checking and finalising component values>PSU goes into current limiting at 1.4Vnot like this
>>3001234the reason you failed is because you triedstop trying and you'll never fail again
>>3001234Oh no, wrong FET pinout. Time to bend some legs I guess. Again.>>3001236not gonna get me that easily, fed
>>3001242This is the second board I’ve had to do this for.
>>3001245Well it's working now, sorta. Removing a backwards electrolytic cap helped. Output voltage and current are regulating with my potentiometers in some regions, and it smoothly transitions from buck to boost. I thought I followed the datasheet recommendation for inductance, but at 100kHz switching frequency my 50uH inductor is maxing out at 0.5A output current, the second half of the current potentiometer turning gives me no increase. The PWM duty-cycle is stopping at about 80%, so maybe I just need to decrease inductance. Once that's done, I'll keep testing, and try the input voltage and current feedback loops.Even with no heat sink, my tests are barely getting these transistors warm, which is encouraging.
why do some MLCCs have 'solder spots' on the sides too? what's that about?
>>3001177I've never seen solder expire, there isn't much to go wrong with it.
>>3001257They are feedthrough/three-terminal capacitors. Works better for filtering because of closer ground coupling
>>3001222>You didn't get them to assemble it for you? Gonna use an oven, hot plate, or hot air?I have a reflow oven
>>3001263thanks! that's crazy, never even heard of such a thing! I wonder how many other things are on these phone boards that I've never realized I know nothing about.
>>3001177>>3001259Flux core can theoretically age, but for hobby use you just...add more flux.
What is up with the price difference here? https://www.digikey.com/en/products/detail/pulse-electronics/HX1188NLT/2265500https://www.lcsc.com/product-detail/C9363.htmlLCSC is usually cheaper but I've never seen a difference as stark as this, $3.00 vs $0.21 at 100pcs. Is it counterfeit or something?
>>3000383C8 looks like it's going to pop
>>3001415
>>3000635>>3000640Thank you! I've actually bought one and it arrived today.I narrowed my choices down to P2 Pro vs P3 and ended up getting P2 Pro because it was cheaper. kekI didn't go for other cameras because I wasn't sure if the resolution would be enough for laptop work. I actually think that a P1 would have done the job to be honest for much less money. I can see components just fine without the macro lens so I assume P1 can't be much different. But I wasn't sure and didn't wanna deal with returns and wait again. Anyway, it's a nice camera. I like it. I can always re-sell it.It's insane how small it is. It's
Got a big tv second hand with a beautiful image. Makes a horrible high-pitched whine only when the display is on; audio with screen off is silent.-Noise localizes to the L301 area on the power board. (Used a paper towel tube and this is the loudest spot)-Pressing directly on L301 produces little or no change.-Slightly flexing the PCB near L301 changes the whine slightly* No visible signs of failure: no bulging capacitors, burns, cracked core, or overheated components.Is it fixable? If I line the underside of the black plastic piece with some electronic silicon sealant might that solve it?
>>3001515The upper cap is bulging and should be replaced first.
>>3001515>no bulging capacitorsmeanwhile the top cap looks like me after receiving a one hour lap dance by Anna Kendrick.
>>3001515
>>3001516>>3001524Oh no. Well I already knew I’m retarded.Would that cause the ringing?Would it be terribly difficult to get a new one and resolder?
>>3001530>Would that cause the ringing?Yes.>Would it be terribly difficult to get a new one and resolder?Not difficult. You can use a cap with a higher voltage rating, but the capacitance value should be the same as the original. Make sure it's oriented in the right polarity or it'll explode.
>>3001532Aluminum electrolytic, low-ESR, radial leads.
>>3001530>Would it be terribly difficultchild's play for myself and others who trained under the Shaolin Master of Solderbut to a noob sporting a dirty tip, total destruction of the TV is inevitabledo not attempt until after watching the videos, acquiring the tools, and practicing for weeks on 200 junk boards
Speaking of coil whine, I recently had an older wireless router "die" after whining for a while. I thought it was something on the board, but it was actually a dying power adapter making something on the router board squeal. There are no electrolytics on the board and the supply itself wasn't making noise. It seems like it works fine again, silently, on a good power supply.
>>3001543The tv is garbage if I can’t fix the whine. And I don’t think anyone would buy it just to fix it themselves.Gotta roll the dice or kick it to the curb. What videos to maximize the likelihood of not fucking it up?
>>3001568>What videos to maximize the likelihood of not fucking it up?start with ''desoldering for beginners''then if you've managed to not ruin the PCB, ''soldering for beginners''personally, i'd skip the desoldering process, so as not to have to remove the PCB(because the more you do, the more likely you'll screw up)instead, i'd cut off the capacitor with sharp side cutters, slowly nibbling around the outsidethen lie down the new cap and solder to the nubs left over from the old capit'll look totally unprofessional, but it's the fastest and safest procedurenot so much Shaolin as Wu-Damn
2 quadrant, non switching, bipolar power supply for electrolysis. +/-10V or so from a 12V supply, +/-2.5A peak output. Will take inputs from a microcontroller DAC for pulsed electroplating.Passive values are not final, and op-amps will be socketed and swappable. Those BA4560s are pretty fast so I may be able to drop that 100pF/100k RC constant to something faster, on the other hand their input offset bias is a little high. I had to use these LMC6482s for the output to get rail-to-rail output capability, but I don't actually think I need them for the sense differential amplifiers, so I may swap those for LM833s (higher CMRR and GBWP). The negative voltage regulator is kinda strange, it's to get headroom for the op-amps to interface with the range of the DACs, for most op-amps these have to be within 3V of either rail. The OP07 for buffering the 2.5V zero reference is silly looking, that probably isn't necessary in the slightest. The CD4053 is to swap between more negative of each error signal, ot more positive of each error signal, to cover both quadrants. I hate how the CD4053 needs level shifting when not used with a 5V (or +/-5V) supply.Suggestions welcome. Feature creep is not.
>>3001574Oh technically I can make it 4 quadrant by adding a passive load resistance in parallel with C3 at the top there, providing a place for power to be sunk. I can add a relay or whatever that switches that later if I want. Without that, to send power back to the rails entails some sort of synchronous buck converter that can act like a boost converter in reverse. Looked into h-bridge buck-boost converters for this, requires a custom control loop and getting that stable was beyond my simulating abilities, let alone my breadboarding abilities.
Hello bros, I need some help on purchasing a soldering iron/station. I have narrowed it down to 3 units that I am interested in for the price range. However, I am fairly uninitiated in the world of soldering -- thus I am not too sure what would be considered a 'quality' clone and straight chinesium garbage. If anyone has some insight on these particular units and how they compare I would appreciate some advice. I am primarily interested in the quecoo brand one, seeing as that one has at least come up in some threads online, but I am open to knowing more about the other units and if they would be better alternatives. Thank you!
>>3001663
i wanted to learn how to solder so i picked up a pinecil and two random wires: I got 1.2mm solder wire with lead (in a random chinese store in my town), and 0.6mm wire, lead free (reputable amazon seller). Everyone told me that wire with lead is much easier to work with so I stuck with that. I've been trying to get a single decent solder on a perfboard with random tht components and I simply couldn't do it, watched dozens of youtube videos, just couldn't do it for days, multiple tries in the 300-350 celsius temp range. Then, I tried switching to the 0.6mm lead free wire, upped the temperature to about 380 celsius and i got near perfect solders each try, smooth like butter.So what's the deal here, is the 1.2mm wire simply too big for the fine pinecil tips, or it may be that its quality is too low, am I doing something wrong or what?Also the pinecil stand seems very flimsy, should i get a decent one?
>>3001672>I simply couldn't do ityou tried the cheapest shity-est solder in the world and it didnt workyou tried a reputable solder and it workedthe mystery kinda solves itself, doesnt it
First time attempting hot plate soldering, apology for crummy phone photoThis was with Sn37/Pb63 solder paste with the hot plate at about 200 C. I noticed that a lot of resistors ended up a bit asymmetrical with a big blob on one side. This is probably due to uneven application of paste with the syringe but is a blob like that indicative of poor wetting/a bad joint?Any other obvious issues that anons can spot?I only did the resistors, being the cheapest components in case I have to throw the whole thing out. I will do the more expensive components by hand afterwards until I get the hang of this hot plate business.
>>3001663Of the ones you’ve shown, the T245 station is definitely the best, and that series of tips also has superior variety available. Assuming it’s actually 180W as it claims. The top one, while being the cheapest, is also perfectly usable, though those helping hands and magnifier are in an irritating position, and it’s firmly no-brand. That said, you should be able to find 70W OLED T12 stations for cheaper than a 180W OLED T245 station, I got mine from KSGER and it had the same handpiece as the first station you posted. The aluminium handpieces are probably better, but superfluously so. If you go to the T12 STM32 custom firmware GitHub repo, you’ll see a comparison of different control boards, some are compatible with the custom firmware, and some have quirks or minor faults you may want to avoid. But the custom firmware isn’t a necessity.It’s also worth being aware that the chinky power supplies inside these soldering stations are never the best. I haven’t had any problems with mine, but some of the high voltage clearances don’t instil me with confidence. If you already have a suitable power supply (24V 3A for T12, 24V 8A for T245, though 19V 6A still gives you a decent amount of power), it’s possible to get an equivalent but cheaper station that plugs in via a DC jack instead of an IEC lead. Something like the Miniware TS1M comes to mind, alongside the regular Quicko and Ksger stations.Whether you go for T12 or T245 cartridge tips, you can buy portable USB irons that take them too, which can be handy for Regardless of what you buy, I’d want to find some properly critical reviews. I bought myself a T12 station almost a decade ago after watching a review, and there’s a lot else out there now.>>3001672The chinky solder probably had no flux, or expired flux.>>3001682Probably fine. You should be able to measure across other connected pads in the board to look for short circuits. Consider how thin a soldering paste stencil is.
i don't have a tech background but can someone help me figure out how class d amps work? class a, b, and ab make sense to me but I don't get the pwm stuff of class d, and how they make sound sound crappy unless they have more tech in them
>>3001708>help me figure out how class d amps workhttps://www.youtube.com/watch?v=4sLk6_qyM24this guy does a lot of blah-blah but at 9m43s shows how varying a PWM duty cycle can be converted to an analog voltage using a simple RC filterso, a class D converts audio to PWM, amplifies it very efficiently (without heat loss), then filters convert it back to audioand more filtering = better sound
>>30016721.2mm solder should work fine, again it probably had no flux (or inconsistent flux, some really shitty ones can have the amount vary through the spool).
>>3001708>how they make sound sound crappy unless they have more tech in themThey don't unless they're legit improperly made, don't fall for the lies of audiophools. A passive filter is generally all that's needed, and anything that doesn't bother spending the couple cents on that is a heap of shit no matter what class it is.
>>3001717awesome ty>>3001730yeah I don't buy a lot of what audiophiles talk about. all reviews say class ab has warmth or some shit and it just sounds like voodoo
Is it worth adding these diodes D5-8 in anti-parallel with the op-amps? These op-amps won't always be in a negative feedback loop since it's a CC/CV circuit, and I'd like to prevent saturation. But I'm starting to think that these diodes won't prevent saturation at all, and won't even speed up recovery time. But the output swing is barely within the rails, and because the non-inverting input moves I can't easily use zener diodes. If there was a large margin between normal use and saturation I could find room for zeners, but I don't think I can.Maybe I just deal with it.
i have pic rel as position sensor for a complete black box.I know the potentiometer is 5k, the voltage is 9.5v and the potentiometer output voltage ~2v in its operating rangeCan i swap this out for a 3 pin hall effect sensor with 0-5v signal without frying the controler?I know nothing about the circuitry of the black box aside from it being short circuit protected and having a calibration procedure
>>3001769Probably, but not by itself. I’d measure the voltage on its pins while it’s plugged in and working, the full range. You can probably get a hall sensor system to work with the right resistors (maybe also an op-amp or two). Knowing the current output of the sensor will also help.Also I see no potentiometer.
>>2997665-DMMA BM786 would be ideal, but get a Aneng AN870 if you want something cheaper, but do keep in mind that besides the fuses, it has no input protection (it will explode if you put it on ohms mode and connect it to 230V). This will be one of your most used devices, so don't skimp out here.-OscilloscopeI'd get a Fnirsi 1014D, it has two channels with a 100MHz bandwidth and a signal generator. It may be chinkshit, but western oscilloscopes with similar specs are much more expensive.-Soldering stationGordak 936A, you can get one with a hot air gun too! The keyword here is 936A, as Gordak is just one of the many chinese companies who mass-produce these cheap but very reliable soldering stations. I've had mine for 5 years and haven't changed one tip in all that time.-Hot air stationSee above, I've hand soldered 0603 SMD resistors without a hot air gun. You just have to use LOTS of colophony resin (don't buy flux)-MicroscopeDon't need one-Bench power supplyGet something with at least 30V 5A and a output switch. Whatever chinkshit you find on aliexpress will be fine for now. Get a linear PSU if you really care about noise.-Hand toolsBuy a set of tweezers and a set of screwdriver bits from your local Dedeman/Home Depot, the brand doesn't matter here. -Wire strippersGet adjustable ones from a hardware store, brand doesn't matter-Board holderYou have two hands -Anything else I'm missingYes, get an LCR meter under the form of a tweezer, like the LCR-ST2. It also has Kelvin clips for more accurate milliohm measurements. I found it very useful since it can test capacitors and inductors at different frequencies. Very cheap for how much it can do.Also, buy or find a pdf online of The Art of Electronics Third Edition. It's a great book that will teach you a little bit of everything in electronics. A easier online-only source which also has plenty of hands-on circuits is "talking about electronics", the site has multiple pdf's which you can download.
>>2998074The DHO is very good, but get the 70MHz version instead. You can very easily hack it to 150MHz (connect it to PC via ethernet cable and run a script).
>>3001828>Also I see no potentiometer.my pic is a potentiometer>I’d measure the voltage on its pins while it’s plugged in and working, the full range.its a 90° pot with 5k ohm and 9.5v supply, obviously the full range is 0 to 9.5v. But in non catastrophic failed operation, the range of motion is capped to at most 60° by a ribbon, the original pot is thus operating between 0 and 6 voltsWhat im wondering is if i should hook a resistor in line with the hall sensor as current limiter? in theory not needed, but the pot is at all times in normal operation at least 1k ohm
Seen a lot of filter + fan diy fume extractors, do they work or am I going to give myself lung cancer?
So I want a laser level for arts and crafts. I can't really justify a 100 dollah one. Browsing eBay I came across picrel. Listing has this info:Wavelength:650nmOutput Power:5mWWorking Voltage:3-5VWorking Temperature:+15 ℃~+35 ℃Laser Shape:Dot/Line(120 degrees as default)/CrossFocusable:YesMaterial:MetalOperating Current:I < 45mADimensions:12x30mm Is this like a LED, where I'd need current limiting resistor? Or is this a complete module where I could whack on 5v from USB A and call it a day?
>>3001896>if i should hook a resistor in line with the hall sensor as current limiter?i'm guessing you realize that- a pot and a hall are gonna have 2 completely different response curves- in fact the hall will have infinitely diff curves depending on placement of magnet- adding a series resistor could alter said curve- this thing you're doing is never ever gonna work, coz you're out of your leaguein any case, current limiting resistors are always a good ideato find the size, you need to determine input resistance of the circuittake a 1.5V battery and connect to the gadget, while measuring current draw(*almost* every one knows that to measure current, you put the meter in series with the battery)input resistance is 1.5V divided by current drawassume you get 1mA, that's 1500 ohmsso you'd like to keep the current limiting resistor to, say, 10% of that so as not to alter curve too much
>>3001896Ideally it’s just a high-impedance voltage sensing input, and it’s a 3-wire pot being fed a low-impedance constant voltage, so a hall sensor outputting a voltage in the same range, maybe with a voltage divider to change the output range, should work.But the self calibration thing you mentioned means it might be doing something else. Is it an electrical self calibration (e.g. measuring the resistance/current to ensure the potentiometer is good), or is it a mechanical self calibration (the machine moves itself to move the sensor through its full range of motion)?As for the linearity of the Hall-effect motion, that’s another matter entirely. There are very linear magnetic rotation sensors that use a diametrically polarised magnetic held directly atop the IC, but I think these usually output digital signals.>>3001941They do work, so long as you have enough airflow and choose the right filter. I suspect you want both HEPA and activated carbon.You could also just run a pipe out a window.>>3001945Current limiting is required if I recall. Look for an equivalent product on Adafruit or Sparkfun to see what they say.
>>3001955Adafruit says "It's a diode with integrated driver." https://www.adafruit.com/product/1057 Which I take to mean it has current limiting build-in.Sparfun doesn't seem to sell lasers.
>>3001691Thanks for the insights, I appreciate it! I decided to pull the trigger on a hakko fx-888dx since there was a good sale on it from a provincial supplier. I have heard plenty regarding that it is a fairly dated (while being an updated variant) and not the most high performance unit available, but rather then deal with the headache of navigating a minefield of chink clones that may or may not be good, I decided to spend a few extra shmeckles to get something that will be reliable and will very likely outlive me while also being capable of any DIY projects I intend to put it through. I hope it all works out, we shall see!
>>3001963>no cartridge tipsngmi
How the FUCK does this oscillator work and why the fuck is mine refusing to oscillate???https://www.instructables.com/The-Ultimate-FM-Transmitter/
>>3001987>mine refusing to oscillateyeah, that'll most definitely oscillate if the parts are good, and the circuit is built correctlytho, most circuits show the cap on the base, C3, going to ground, not Vccbest guess is it's outside the FM range, so you cant find ityou should be looking for a spot on the dial where the background noise disappearsyou prob wont hear audio coz the mic is not amplified enough, or is the wrong typemaybe inject music from phone, or whatever
>>3001867>-Microscope>Don't need onebro, what kind of a vision do you possess when you don't need a microscope in 2026?wtf do you even work on? old tube radios from 1930s by any chance?
>>3002057Literally anything larger than 0402.
>>3001968Decided to cancel hakko order and swap for something a bit more modern and better performing. Cheers
>>3002066Given that an average age of a 4channer is over 35, I gotta say I'm impressed.I can't see shit anymore.>>3002069>PACEhow much do these go for? that can't be cheap.
>>3002071After taxes it was a little over $550CAD. Included with a 3 tip bundle with the 1/32" sharp conical, 1/32" chisel and 1/16" chisel.
>>3002073but why bro? why go with that instead of a much cheaper C245 JBC clone from China?
>>3002074Because I want to have something I can depend on, and will offer a lifetime of good performance. I understand the chink clones are much better now than they were even a few years ago, but I just wanted some peace of mind and got something I know will be good and reliable.
>>3002077I have two soldering stations and my main one is a JBC clone and I haven't had any issues since I've started using it 3 years ago. It probably runs circles around that PACE too.... for 1/5th of the price.
>>3002077I don't really know why people think soldering irons will be unreliable, especially for something you're probably going to use rarely. I've got a 30 dollar chink station that works fine, after a decade the cord tore where it plugs into the body and instead of spending 10 bucks on a new handle I just bought a pinecil instead, which also works great. It's a heating element in a tube, these are not precision pieces of equipment.
>revise PCB design to fix a problem>a different part of the PCB stops working in version 2Why didn't you guys tell me to back up my previous revisions in KiCadIt's not good enough anons, I expected better of you
anyone got schematics, even block diagrams, for very early chinese inverter type welders?I got one that has died years ago and may be 2 decades old or so. I was in there once and didn't really feel like it, convoluted, analogue, discrete mess.I'll have another go at it and if not get some goodies out of there.
>>3002119here's the sucker. I remember a few things about it:I have no recollection of it failing only to find it dysfunctional one day.It's built lile a tank especially in comparison to modern units. Sure this is partly becauae of bad design and discretes and all that but I also remember it running way above rated duty for days on end without issue.It's an absolute PITA to take apart and doesn't get any better inside.The transistors are IGBT, there's a fat transformer, heatsink etc. should I want ro gut it.But first I want ro try find the fault and fix. I try to document it here, partially for myself to be a bit more organized than usual. Would be great if anyone else finds this thing fun and want's to remotely participate.I'll be pulling the case...
>>3002074NTA but my chinky T12 station blew a power MOSFET. I could fix it because I owned another iron, and some russian had reverse engineered the schematic. I’d rather own something that breaks ten times as often if I can fix it in under an hour when it breaks, but not everyone feels the same.Also some of those chinky stations are bad for leakage current, there’s an autistic open source soldering station project you can build that specifically addresses that.>>3002116I think there’s some sort of version control in KiCAD now? If not, there’s a GitHub plugin.>>3002122Obvious possibilities:>thermal fuse in power magnetics>regular fuse>dead power transistor (maybe because of dries out thermal grease)>cracked PCB solder joint on heavy components>overheated cracked power resistorGood luck. Should be possible to reverse engineer the general circuit flow, and derive the fault through some simple troubleshooting methods.
>>3002116Dude, we shouldn't have to tell you to save previous iterations. That's basic project management.
>>3002122Since I cant exactly remember what the fault even wasmaybe I should of powered it on first KEKI can still so that laterSo. this is presumably the conteol board as most of the controls are wired into it and it's shielded on top of it all. There also seems to be some power supplies which is surprising, given that there is a big ass 220V xformer (top right, input filter top left) that puts out 9, 16, 17, 18, 20 and 48 V and easily 100W+. The board seems to also function as power distribution board, presumably the relays switch when operating conditions are within limits and power other boards. As promised lot's of analog logic, yes those are mostly OPs and stuff.Logic is labelled WSME(?)200C5, its probably specific to 200A welders.
Naturally there are some transistors. Output transformer, output inductor with something inside that is not a current sensor IMO because that's on the positive. On first glance there's a surprising lack of driver circuitry close to the power trannys.
>>3002128I believe the weird thing in the output inductor has something to do with coupming the HF start to the output. Here's the other side. 4 more transistors on the other half of the heatsink, beeeefy tracks, another inductor, so it seems, on the positive, current sensor, board for HF start with flyback, a board with bridge rectifiers and huge power resistors in the back (???).The whole thing is suprisingly self-documenting for todays standards, no etched chips, clear markings on silkscreen.What should I try to pull first? The whole thing still looks nightmarish, random wires, glue on everythingrigid conductors soldered on with a 1 kw iron o algo
>>3002131The odd board with rectifiers appears to be more power stuff, I suspect it connects directly to the transformer providing many voltages, rectify and regulate (5 lin regs, 12 and 15V) interestingly 2 classic half bridges (TIP41/42C) driven by an 8050 each, optos, schmitt triggers, 555 timer and some OP.All visually okay.
>>3002135I forgot it's labeled: WSE-160A (!? so for a smaller machine?) and no power resistors or any components on the back. The resistors are mounted on a heatsink behind the board.Logic board / mainboard. I dont suspect it but pulled it, interesting how it works. Looks. like repurposed transformers on there. I might buzz it out, look up unknown ICs and check where the connectors go later.This is labelled WSME200C5 as mentioned earlier. Production date is 20110622.>>3002124>chinese vottage industry thing>fusesure thing
>>3002126>>3002116 (me)I made this shitpost out of frustration, but the update is that I was able to spend an hour talking the problem through with ChatGPT and it led me to the solution, knowing nothing about my layout apart from the data sheet for the IC.This might more of a poor reflection on my own skills rather than the power of the AI, but I was pretty impressed that it could give me a sequence of logical steps to try, could tell me when an observed signal didn't seem correct, etc.Turned out to be a resistor that looked soldered but had no connection.
>>3002141>looked soldered but had no connection.operator errorwho coulda seen that coming?
>>3002140Pulled the board with the obvious full bridge. Finally something that is quick to check.Full bridge, 600V 60A IGBTs, RD snubbers (4007 of all things, KEK I wonder how slow they run this thing or if the snubbers are even doing anything), two full rectifiers, seemingly connected parallel and mains connected via relay charge the caps. Freewheeling diodes are ultrafast ones (black heatsinks).Notice the bulging component on one of the IGBTs heatsinks, perhaps a temperature probe.So it works as expected. Full bridge drives xformer. Split supply is taken from the secondary, diodes are on the output board. The 'positive' terminal is a center tap on the secondary. Two groups of switches, so a half bridge, control the current and direction through the output inductor. The power board seems to buzz out okay so far, no shorts that shouldn't be or so.It really pains you just to think of how much power is wasted in those boards and how much space one could have aaved with a few simple ICs or even MCUs. I know Uncs created designs like. this - because they had to. But 2011 isn't that long ago. I scorched the board btw.
>>3002178It looks like the temperature protection switch comes prefucked from the factory, according to sales pictures I found online. I'll assume it's fine for now.
>>3002179>switch comes prefuckedusing factory rejects for safety related parts?how heebish of them
>>3002179Wouldn’t that be an auto-resetting one? Only the manual reset ones have a button sticking out of them.
>>3002185Yeah I'm sure mine is auto resetting as no one is supposed to every touch anywhere near there. I mean it's on a heatsink that is directly on the collector of a Transistor that is part of a full bridge for rectified mains.>>3002184I dont think they are rejects. Thats just how they come. The flimsy metal colar can be expected to be pulled straight when mounting anyways.
>>3002179So I'm just testing my way through the power electronics bit by bit as that's the trivial part and also not worth fixing.Multimeter needs to charge but so far, all diodes work, relays work, mosfets work and VCE is as advertised...This got me thinking: Am I mistaken if I believe it's not that hard to simply computerize the whole thing?Feed 10 knobs, 3 buttons and a few sensors (temp, current etc.) into a MCU. Outputs are mostly relays, gate drive, valves, indicators and display. Alot of the function is merely a sequence, button push - > gas valve - > HF ignition - > low current arc - > button released - > ramp up to set current... etc.Judging be the circuitry in there the power part is probably pwm with several envelope generators that are factored in. Should be easy to do in SW.Anyone?
>>3002131It's quite weord and I think this board is not exactly what I thought. I thought it is a simple full bridge serving the primary. Bit the transformer has 2 seperate primaries and if one does look closely this board has 4 connections to the xformer, 2 red and 2 yellow leads. IIRC each or the primaries has one red and one yellow lead. This would have to mean the 2 transistors on either side serve one coil and the coils are connected in the opposite direction to each other. I see nothing wrong with it, one coil can reset the flux from the other coil. I'm just not sure why.
>>3002194Yeah it’s doable. First hurdle is reverse engineering the existing circuit. Second hurdle is getting nuked by EMI.Honestly I’d keep it as is, because an analog circuit (that at some point worked as designed) is far easier to troubleshoot. But with an MCU you’d be able to do fancier stuff, like custom pulsed modes and custom controls.>>3002196Maybe they’re wired differently depending on the desired output voltage? Serial vs parallel. Or equivalently for 220V vs 110V. I expect that second reason is the purpose behind the dual bridge rectifiers.
>>3002196I'm at a total loss! >Picrel for clarification.Someone please provide insight.This thing IS a double two switch forward!WHY?It's not the benefit of doubling V_CE, the DC link cap is rated 400V and one transistor alone is rated 600V.Why double it up or why use 4 switches?Using 2 opposing primary windings one could make the transformer reset using only 2 low side switches, just like a single 2 switch forward would need only one primary and 2 switches to do everything.The only benefit I can see, but this can equally be had with a simple 2 switch forward: This resets faster when the other winding comes on and allows for high duty cycle. I'll give them that. Are they doing some black magic with the control scheme? Is it so they get a particular waveform out that saves them adding much output capacitance or something?Confusing.
>>3002198Sorry for the spam but can someone sanity check my theory:It has to do with AC mode. In short: A welder like this can provide AC at the output, for this AC mode is chosen AND an AC balance is set. This can be something like 70% pos 30% neg. I think it has to do with this. In DC mode only one two switch forward operates, and resets as is typical for the topology. This means the following inverter stage sees voltage pulses < 50% duty and presumably the output inductor makea that 'DCish'.Only for AC one two switch converter comes on as soon as the other turns off and for as long as the AC balance setting demands, which probably cancels and reverses flux fast (?) but any remaining flux (AC balance) is reset in the usual way and the following stage sees alternating output on the transformer and can drive the reverse polarity group of switches when this appears on the transformer and vica versa.
>>3002200It's not spam because your posts are relevant to the thread topic, the thread is past bump limit, and this board is slower than molasses in the dead of winter.
>>3002200If it’s an AC welder, it needs to have something more complicated than just a bridge rectiifier on its output. I assumed this would be done with transistors on the secondary side, but I guess it’s probably possible to do it with a raw transformer’s output instead.
>>3002246Yes, it's an inverter.
besides pliers, are there any other improvised holders/3rd hand you can think of for soldering?pic related looks neat but have not tried it.
>>3002345Vice grips, an actual vice, croc clips, or the final boss which is just realising that 3rd hands are gimmicks and the easiest thing to do is just put one thing on an overhang, bring the other thing to it, and add the solder with the other 3 free fingers on your hand.
Another veroboard project. More or less this >>3001574. Being able to route out single sided boards would actually make this a lot faster and more compact, I should probably revive my old Snapmaker 3-in-1.
>>3002358Image has been forgotten, once again.
>>3002359I don't want to insult but I believe you did not take the time. needed to work on your layout and instead went right to fabrication. You could have made this alot more compact.Picrel here was made in a very similar technique and didn't either receive much consideration but yet it appears more space efficient already.
>>3002367>resistors in an IC socket
>>3002370Smart. It's a prototype.
>>3002371>prototypefor a prototype, use DIP headersstill easy to replace resistorsbut no iffy contacts
>>3002381fuck you it's what I had
>>3002381That would work too.>>3002382It's a good idea, Anon. You could probably make a bunch of circuit templates that way so you don't have to build everything from scratch every time you put something together. It's modular perfbread?
>>3002384You can even solder components into the sockets and they simply come cheap as they arent obscure.Idk I only have the picture from the top but perhaps I can check if the thing is in the box of things. It's likely that it gets worse on the backside. I often wire wrap everything on perfboard.Regarding the modular approach: I'm a big fan of that. For example I have made boards for standard components such as TL494 or using various OPs such things as peak detectors, waveform to waveform converters, U to I and vice versa etc. things, which certainly are way bulkier and maybe fiddly sometimes but if you think it through well and give the things a few pots, jumpers and perhaps places to add or change fixed components you can work very fast on some problems in the future.>>3002200So I worked my way through and I am relieved to see I have a complete, seemingly faultless power section. From mains to the output. So worst case I can rig new logic. Bad news about this is that from here testing will be quite a bit more involved and dangerous too, I dont have an isolation transformer that can handle even the logic section, judging by how no fucks were given about energy efficiency. It also requires a fair bit more understanding.I think the best way to proceed would be to put it all back together and take mains off the rectifier for the pulse transformer stage, then check all obvious voltages and see if there is gate drive. But this involves lots of putting back together and still leads to some ambiguity as I am not entirely sure if the logic is dumb dumb or if it has some basic sanity checks, like voltages and currents here and there.I'm also considering spending more time reversing, powering sections on the logic board and faking required inputs to see if it generates anything.Might also look at the few things I haven't looked at yet. One more mystery board and the HF start generator. Other approaches?
>>3002389This bit is still a bit mysterious. Looks like a massive RCD snubber. There's some diodes, maybe some serious transistors but I haven't looked them up, the things at least don't look like they receive control signals but might be diode connected o algo. There is a massive amount of high quality capacitance, the fat resistors in the background but the blue green and white twisted pairs go off to the previously mentioned bank of 6 50W 100R. All sounds like a RCD snubber. Only that it hardly has any obvious connection for the implied power. One connection goes to the pulse transformer. And it's connected to the logic. Note the two relays and a small TO-92 package.I get that a good amount of snubbing is needed, especially since the impedance at the output might suddenly go from 0 to mega.So why is this thing apparently active in some way? If you need a snubber it's not like you want a logic that can disconnect it.Could it be that they can just extract the energy through the transformer and for that purpose use a higher voltage winding, explaining the seemingly flimsy connections for such a beefy snubber?Other thoughts?
>>3002396Maybe it has something to do with hot start. The current is higher when the arc is struck and decreases immediately after beginning the weld. Arc force is about maintaining voltage throughout the weld.
>>3002397>maintaining voltage throughout the weldWYM? TIG or stick, the voltage will always be a function of resistance, which is dictated by gap length, ionization, material, its cross section and length to earth lead.>The current is higher when the arc is struckNot really, or depends. If TIG and HF start the current is virtually zero, the gap ionizes under negligible AC current and the inverter ramps the current up according to the programmed envelope.Either way I don't see how 'not working' is related to current at the start of a weld or voltage being maintained throughout it.>>3002135>>3002140Got on with those two suckers. The plentyfull secondaries on the auxiliary transformer have not so much to do with voltage requirements but rather most sections on the logic being galvamically isolated and talking through optos only.The two transformers that I suspected to be reclaimed PSU parts used for PSU stuff... still look very much like reclaimed PSU stuff but at least one is used for the gate drive in the pulse transformer 'not full bridge'.The smaller board is one beefy boy of a gate driver for the output IGBTs. They even went for beefier 200 mil (?) pitch connectors to hook the output IGBTs up. I guessed before and now even more so believe those run much faster than the ones for the pulse transformer. The board seems to pull the base to either +15/-12 V using a TIP41/42 half bridge for each channel. Again this is why so many secondaries on the auxiliary transformer. Positive voltage regulator for low and high voltage. Some more logic, this board makes the decision regarding DC/AC operation and accordingly ignores the AC choice when stick welding. It takes the inputs directly from the front panel, including AC balance and AC frequency. Relays controlled by the main logic. Unless they are used to switch the signal I can not tell where it takes a signal from the main logic. It must take at least a signal proportional to the current error or something.
Is this the right sort of thing for adding a bodge wire on a slightly cramped surface mount board? I heard there is also "kynar insulated" wire but it sounds like I want the enameled stuff where you can burn off the insulation with the iron.I didn't fuck up my own circuit board design, but my friend did, and so he needs to put a bodge wire on his board. My own board is fine and has no errors by the way
>>3002367Dot-board or strip-board? I’ve acted to minimise connecting wires.
>>3002442>I heard enameled ...This is not always the case. Make sure it says it's self fluxing or solder-able. Wire like this can be a real pain in the ass. Veropen ships with nice wire. Transformer, motor and magnet wire is wrong and needs mechanical removal of the enamel. Sometimes you can get it to work with a very high iron, lots of flux and patience, or by forming hooks and sticking into the hot air gun, leaving them in to bake until you need.>KynarGet it if you can. Got some rolls here. Kynar and some PVC insulated ones are really nice as you just touch the cut end to an iron and the insulation will simply shrink a mm or so back, exposing a real fine tip.>>3002428It's me, poasting from the computer. I might have gotten lucky and it may simply be that the driver board has a short. Sorry for how picrel looks but, red and blue, coming from the regulator behind the rectifier, seem to have 0 R and 0 V between them. This shorts out most active components. I could try the good old loading the short and sniffing for funny spots. Randomly de-solder stuff or break tracks or think of a reason explaining the measurement but honestly 0R between Vcc and GND ? I dont like thinking.>>3002444Dot or stripDot. And another similar one (found em) on bare perf.>Minimise wiresWhy? And I believe you can still compress that without too many wires if you put thought in but maybe im wrong.
>>3002454Gentlemen! Place your bets.Is it an IC? Which one? The linear regulator? Any of the passives? Debry?It's probably not debry. The whole thing is filled with metallic grinding dust so I cleaned it quick and it didn't help.So, suggest something and I will de-solder it and reveal to you whether you were right or not.>Image limit
>>3002460I'ts not the lin reg.No point in guessing the 555 it comes out next.
Image limit reached. New thread >>3002473
>>3002345I've heard some people use blu tack but I always forget to try it.
>>3002487Blu tack is pretty good but the heat can make it stick onto whatever it is you're trying to hold on
>>3002487>>3002719yeah, it can be messy but blutack is not something you'd have around your home since it's used for temporarily attaching things.
>>3002460I got a feeling its the relay. Didn't read any other posts.
>>3003532already answered on the new thread>>3002477>>3002473>>3002477>>3002473>>3002477>>3002473>>3002477>>3002473
