|
Shame Boy posted:You can absolutely get 5V relays that make decent clicks, but the thing is they're gonna be drawing like 100-200mA a pop (lower voltage = higher current, every relay basically does the same job and so uses the same amount of power after all), so if you have 64 of them that's nearly 13 amps. Go up to 12 or 24V and it becomes much more reasonable, a quick search for a 24V relay datasheet says its particular coil current is only ~25mA, and 24V is still well within safe, easy to get power supply voltages that you can switch with cheap, easily available MOSFETs. Yeah, I love it too, and I think it would give the thing a cool feel that I couldn't get from speakers. This sounds like a job for an XBox power supply: 5V 1A and 12V 18A. A high-power cascading shift register/current sink like the TPIC6B595 would let me bit-bang relay outputs, and if I latch every time I send a new bit, would just naturally have that cool "wave of relays clacking" effect without any shenanigans on the microcontroller to emulate it. Okay, thanks! I'm off to source some parts!
|
#
?
Mar 31, 2023 17:27
|
|
|
|
| # ? May 29, 2024 18:33 |
|
|
The low voltage, high current power supplies would come with additional problems that I'd recommend avoiding. Like wires getting hot. A 24v power supply for the relays would be ideal, and either a step down power supply for your microcontroller, or even a separate 5 or 3.3V one for your semiconductors.
|
|
#
?
Mar 31, 2023 18:12
|
|
|
ante posted:The low voltage, high current power supplies would come with additional problems that I'd recommend avoiding. Like wires getting hot. Okay, thanks. And, of course, this is going to require me designing my own PCB, because all the existing "LED Driver breakouts" handle 0-17V.
|
|
#
?
Mar 31, 2023 18:23
|
|
|
cruft posted:Okay, thanks. And, of course, this is going to require me designing my own PCB, because all the existing "LED Driver breakouts" handle 0-17V. I mean there's tons of "relay driver" breakout boards if you don't want to make your own, but it's not a particularly hard circuit either and would be a pretty good first project. You basically want something like this:  Q1 needs to be a "logic-level" MOSFET, either they'll straight up say they're logic-level or they'll have a gate threshhold voltage (the voltage at which they start turning on) a bit lower than 3.3V, like maybe 2-2.5V (lower is better). This isn't going to be sinking huge amounts of current, so it doesn't have to be too fancy a transistor or anything. R1 is just to keep the transistor from turning on due to stray voltages before the microcontroller has started driving the pin. D1 is there because when a relay (or any kinda inductive thing, for that matter) switches off suddenly, it generates a big voltage spike as its magnetic field collapses. If you don't handle that, it can fry the transistor, so you usually put a diode in parallel like this so the spike can just short itself out. Something cheap and easy to get like a 1N4007 would work fine. Duplicate this 64 times and there you go 
Shame Boy fucked around with this message at 18:51 on Mar 31, 2023 |
|
#
?
Mar 31, 2023 18:49
|
|
|
Shame Boy posted:Duplicate this 64 times and there you go LOL, maybe now it makes sense why I'm looking at current-sinking shift registers  Shame Boy posted:D1 is there because when a relay (or any kinda inductive thing, for that matter) switches off suddenly, it generates a big voltage spike as its magnetic field collapses. If you don't handle that, it can fry the transistor, so you usually put a diode in parallel like this so the spike can just short itself out. Something cheap and easy to get like a 1N4007 would work fine. This, however, is not something I knew about, and I will probably be posting a PCB design in a bit here for people to look over and say "dude you are going to fry your microcontroller with that" or whatnot. Thank you!
|
|
#
?
Mar 31, 2023 18:54
|
|
|
Shame Boy posted:Honestly the speaker idea makes a lot more sense and would prolly be easier to do, but I really love the idea of a big grid of relays ch-chunking like it's some 1940's telephone exchange sooo Saw a guy at the Vintage Computer Fair one year that had built his own relay-based computer, and had one clacking away at his little stall. It sounded great, even if it was impractical for all but the most basic of computations. Looked great too as it had lots of LEDs to show state/etc. He said he'd actually built a bunch of them to try different relays to get the best sound, and his favourite were a specific Eastern European brand that was quite a bit louder than the rest.
|
|
#
?
Mar 31, 2023 19:23
|
|
|
cruft posted:LOL, maybe now it makes sense why I'm looking at current-sinking shift registers A typical IO expander interface is a SPI or I2C input where you write a new full word of bits and then the parallel outputs on the chip all change directly to the new values after the full word is received. There's still a shift register in the SPI receiver implementation, but the parallel outputs don't have shift behavior or memory
|
|
#
?
Mar 31, 2023 21:41
|
|
|
May I trouble you guys for some advice? I've taken apart a hoverboard, and I want to repurpose it and I think it is mostly hinging on one factor right now. Are the motors in the wheels? They are, aren't they? See pics:  Thanks, any input appreciated
|
|
#
?
Mar 31, 2023 21:54
|
|
|
Yes. Brushless motors. I can't make out in the picture if it uses sensors or not. Does the motor cable have three larger wires and 5 smaller wires?
|
|
#
?
Mar 31, 2023 22:01
|
|
|
CopperHound posted:Yes. Brushless motors. I can't make out in the picture if it uses sensors or not. Does the motor cable have three larger wires and 5 smaller wires? Wow, this thread, you guys rock. Look, here is a pic and then a closeup, and yes, they come out bundled and in the closeup you can see where they break out. 5 small and 3 larger.   I don't need the lights for this thing to work. I want to mount the wheels on a separate armature, probably made from scrap lumber and I'm probably going to extend the cables running to the motors, run a steel rod or pipe through the body of the hoverboard and mount it so I can control it with my feet, as if I were actually riding one, but using it to control the motion of the motors for something else. I think. Does this sound doable so far?
|
|
#
?
Mar 31, 2023 22:24
|
|
|
yeah it does but do you actually want to stand on it? you could make a lever or many other control methods just as easily. what are you actually making?
|
|
#
?
Mar 31, 2023 22:34
|
|
|
I don't want to actually stand on it, I want to sit in a chair and use it on the floor in front of me, just for clarification. And I'm working on a mechanical sequencer to play music, or an electric prayer wheel, as it were. Here's a 14 sec video of what I've gotten together over the past few days: https://www.youtube.com/watch?v=35YMfe0JHQM It's got an optical sensor that can perceive depth, and I'm rotating spoked wheels in front of it with bottle caps stuck on the spokes because I'm out of baseball cards. It's converting the positional info about the bottle caps to simple midi notes, which I'm sending to a small synthesizer in that pic. the fixed distances of those bottle caps can be changed, you build sound patterns in real time using the arrangements of simple physical objects, in this case bottle caps. But at present I don't have any precise control over the RPMs and so I can't really integrate it that much with my other gear. But also, the speed and finesse that you can control a hoverboard might be really useful with this. I'm thinking of mounting some tachometer or something, not sure what. Oh, and a big part of the draw for this thing was that you can reverse either wheel as well, that's not an uncommon feature in electronic sequencers. But I would mount the wheels onto some wooden armature that I can easily engage/disengage and sit in front of it, perhaps playing a different instrument with my hands. petit choux fucked around with this message at 23:50 on Mar 31, 2023 |
|
#
?
Mar 31, 2023 22:57
|
|
|
Foxfire_ posted:The category of thing you'll want to search for is "IO expander". Shift register would also work, but it kind of implies the bucket brigade type memory and logic in the outputs. I think a shift register is perfectly acceptable. The 74HC595 from memory, I think, was the one with the separate latch register. Might be a little cheaper. But both will work, no big deal either way
|
|
#
?
Mar 31, 2023 23:03
|
|
|
petit choux posted:I don't want to actually stand on it, I want to sit in a chair and use it on the floor in front of me, just for clarification. And I'm working on a mechanical sequencer to play music, or an electric prayer wheel, as it were. Here's a 14 sec video of what I've gotten together over the past few days: sounds like a pretty good use case then, it looks like each hovorboard motor has it's own tilt sensor so it should be easy-peasy to control them independently from a petal setup
|
|
#
?
Mar 31, 2023 23:32
|
|
|
PokeJoe posted:sounds like a pretty good use case then, it looks like each hovorboard motor has it's own tilt sensor so it should be easy-peasy to control them independently from a petal setup Well that's great to hear, I was under the mistaken impression last night that the motor shell was integral to the body of the hoverboard, which would be more of a PITA. This isn't going to really address the RPM issue but it should be fun anyway. Ultimately I'll want fairly powerful yet quiet motors that can reverse on a dime and I can control the speed with precision. This is probably not it, but the ease with which it all seems to be coming together makes it feel right at least for fun. petit choux fucked around with this message at 00:18 on Apr 1, 2023 |
|
#
?
Mar 31, 2023 23:44
|
|
|
Now I've never torn one of these apart, but my guess is that these are the tilt sensors and the other bigger boards are the motor/charge controllers. You should be able to mount those sensors to a petal setup of your liking, and the tilt sends a signal to the motor controller board to control motor so the plastic case is likely just a case. It should be easy enough to test, gently pop the little circled board off those mounting pegs and tilt it to see if it makes the motor move when powered. PokeJoe fucked around with this message at 23:51 on Mar 31, 2023 |
|
#
?
Mar 31, 2023 23:48
|
|
|
PokeJoe posted:Now I've never torn one of these apart, but my guess is that these are the tilt sensors and the other bigger boards are the motor/charge controllers. You should be able to mount those sensors to a petal setup of your liking, and the tilt sends a signal to the motor controller board to control motor so the plastic case is likely just a case. I thought so but they appear to be nothing but lights. Here's a closeup:  but if I'm mistaken please let me know because that would change everything. I think I've located the model of this one and they are shown with lights in that location, in addition to pretty elaborate lights on the wheels. If the tilt sensors aren't on the big board I'm not sure where they would be. But the underside of that board has some interesting stuff going on that I haven't taken apart yet. This is also like a really cheap model and runs on lead acid and has very little metal inside it, I'm now discovering. Looking at the good ones that cost about 3x what this one does I'm seeing some real sturdy hardware. I did visit the manufacturer's website and it appears the current setup may be more like that, they appear to say there's a gyro chip in that location. But having removed it, I'm not seeing anything but these LEDs on that board. That is on the low end model, I don't think all the parts I saw are relevant to this one. petit choux fucked around with this message at 00:16 on Apr 1, 2023 |
|
#
?
Apr 1, 2023 00:05
|
|
|
Oh but here is a question. I believe I am correct in saying it has 3 phase power or whatever, I have a battery powered amplifier like that, you are connecting to 3 points on 2 batteries, it appears. So how may I determine what voltage wall wart I would replace those with?
|
|
#
?
Apr 1, 2023 00:22
|
|
|
Oh yep those are definitely surface mount leds. The sensor is likely on the control board then
|
|
#
?
Apr 1, 2023 00:24
|
|
|
petit choux posted:So how may I determine what voltage wall wart I would replace those with? The physical motors themselves will probably work in the 18-52v range but that will depend on the motor driver.
|
|
#
?
Apr 1, 2023 00:43
|
|
|
Can any PCB wizard tell me why this Aliexpress-special melted down? Front of board:  Back of board (notice that greasy streak? that was caused by the heat, I think?):  When I plugged it into a fancy laptop panel (https://www.panelook.com/B173ZAN06.8_AUO_17.3_LCM_overview_55996.html), and into a USB-C power supply, and into my GPU, the panel lit up and showed my desktop, and I was actually getting 4K 120hz for about 30 seconds, then it became scrambled and began power cycling. I could feel heat radiating from the board while I was holding it from the sides. I touched the back of the board and it was skin-burningly hot. I touched around that area on the front of the board below and could only describe it as thermonuclear. Like my thumb still feels burned an hour later:  What's more, I tried to place a tiny Raspberry Pi self-adhesive heatsink on the R2R inductor, and the heat was making it slide off. I let it cool and tried to gently twist it off to reposition it, and the 2R2 inductor cracked in half! As if it were super brittle. I do have one of those solder rework heatguns if I need to replace that inductor, but is there maybe a more efficient and durable version? Also, does anyone know what that unpopulated block is for? I was hoping the board might have some audio line-out capability. edit: The two components around the mega-hot area are an SS34 Schottky Barrier Diode, and a 2R2 Inductor. I really wish I could find my infrared camera smartphone dongle. Zero VGS fucked around with this message at 01:05 on Apr 1, 2023 |
|
#
?
Apr 1, 2023 00:56
|
|
|
CopperHound posted:Motors have some complicated electrical physics stuff going on, but roughly top speed depends on voltage and torque depends on current. Judging by the picture something around 24 volts is what it is running at right now. Do you have the original charging brick to check? No, I'm afraid I don't have the original charger. I do have a pair of 9v batteries I could perhaps drop in in place of the others - one of them was leaking acid when I dismantled it, no power at all. Would I be all right to put in the other batteries (they're all charged up)? And if so, would I be able to use my multimeter to determine this, assuming it all works at that point? I would love the option of not using these batteries perhaps.
|
|
#
?
Apr 1, 2023 02:25
|
|
|
Zero VGS posted:Can any PCB wizard tell me why this Aliexpress-special melted down? My first guess is that that's an inductor and diode for an async boost regulator whose output is shorted (maybe because your LCD failed, maybe because the output capacitor C11 on the board next to it failed short). Short circuit protection is a fancy feature for async boosts so they usually don't do it, especially on cheap products. The good news is that shorting the output won't necessarily damage the boost regulator IC itself since it only stresses the diode and inductor, not the MOSFET. quote:Also, does anyone know what that unpopulated block is for? I was hoping the board might have some audio line-out capability. edit: Also hard to say without tracing it out on the board but there's one connector populated with USB signal names labeled and another unpopulated that also has 4 pins. Might just be an alternate location for that USB connector (the header, not the type C). I don't know of any IC with only 8 pins that can get audio out of HDMI. Looks like it's just a power supply board that passes the HDMI signals straight through. Stack Machine fucked around with this message at 04:51 on Apr 1, 2023 |
|
#
?
Apr 1, 2023 04:15
|
|
|
 work in progress 300W Power Converter 9-36V in 8-24V @ 20A out (300W max) I'm working on a new project: a rugged 300W DC-DC step-up/step-down power supply for working with cars and power tool batteries. Basically, I want to run high-power automotive stuff like tire inflators and my ham radio using a Milwaukee M18 chainsaw battery. I'd also like to use the M18 battery to recharge my car battery in an emergency (it's not powerful enough to turn the starter, but it can put a superficial charge on a dead car battery). Features: Digital output voltage control. ADCs measure input and output voltage, output current. Power, Charge, Energy monitoring on output. Robust to electrical faults, incorrect hookup. Audible alarm for reverse polarity, overvoltage. Temperature sensors. 5V 1A housekeeping power supply. 100% controlled over I2C. Core chip: LT8390 60V Synchronous 4-Switch Buck-Boost Controller TDK i7C2W020A120V-003-R is a power supply module based on the LT8390. It's very similar to the circuit on pg 28 of the LT8390 datasheet. This is the heart of my device. The input voltage passes through a fuse and a pair of big back-to-back FETs controlled by a LTC4365 Overvoltage, Undervoltage and Reverse Supply Protection Controller. It should survive +100V overvoltage and -40V reverse polarity. The output is also protected with back-to-back FETs, though under GPIO control, so they will not have as fast a reaction time. But they will prevent sparking when connecting/disconnecting cables, because they isolate the large output capacitors from the output connector until the microcontroller turns it on. A LTC2946 Wide Range I2C Power, Charge and Energy Monitor sits on the output, with its own current shunt downstream of the output filter caps. It's a DC-DC swiss army knife, designed to be controlled by an Arduino.
|
|
#
?
Apr 3, 2023 00:22
|
|
|
I've used a TDKi7 module before. Make sure your heat sinking is good
|
|
#
?
Apr 3, 2023 04:11
|
|
|
ante posted:I've used a TDKi7 module before. Make sure your heat sinking is good Heatsink? I was just going to blow on it with a little fan. I will have a temp sensor glued to the TDK module, so I can monitor the temp in operation. (Is this good enough? I don't see provision for a real heatsink.)  There's also this ominous footnote in the TDK docs:  I think this has something to do with the LT8390's loop compensation (see LT8390 pg 25). I guess it depends on the values of the Rs and Cs TDK put on the LT8390 Vc pin? I was planning on putting 1600 uF of aluminum polymer caps on the output, but maybe that is a bad idea? TDK support didn't respond to an email, but maybe I haven't bugged them enough. ryanrs fucked around with this message at 05:02 on Apr 3, 2023 |
|
#
?
Apr 3, 2023 05:00
|
|
|
Fan's great. Should work fine. Also, you should be fine with that load, though it is massive
|
|
#
?
Apr 3, 2023 05:55
|
|
|
Speaking of massive loads, it'll be an opportunity to try out an active load I built a couple years ago. I designed it in EAGLE + python scripts to gently caress around with polar coordinates. Historical context: At the time, people were making fun of Pence for saying 'lodestar'
|
|
#
?
Apr 3, 2023 07:25
|
|
|
Stack Machine posted:My first guess is that that's an inductor and diode for an async boost regulator whose output is shorted (maybe because your LCD failed, maybe because the output capacitor C11 on the board next to it failed short). Short circuit protection is a fancy feature for async boosts so they usually don't do it, especially on cheap products. The good news is that shorting the output won't necessarily damage the boost regulator IC itself since it only stresses the diode and inductor, not the MOSFET. Thank you for the insights. I have a basic multimeter with no capacitance testing, but it has continuity, and each of the capacitors is grounded on one side and not the other, so I guess that means those aren't shorted? The USB on the board is claimed to be for a touch screen; I guess if your display panel has a digitizer, the signal would go over the 40-pin EDP into that board, then out through that USB header? As for the super-heating diode and inductor, do you suppose I could upgrade those with components from Mouser that run more efficiently, or at least have better heat dissipation? Maybe use components with leads, so they're not transmitting so much heat to the PCB? Can I run multiple of that SS34 diode in parallel to spread the current, without anything catastrophic happening?
|
|
#
?
Apr 3, 2023 10:34
|
|
|
So some more adventures with my laptop, though this time not my fault think Some F and number keys stopped working on the keyboard and since it doesn't seem to be easily replaceable, I'm trying to see what's up. You can see there's a weird white spot on the second trace from the right, which I thought might be an issue. However I busted out my cheapo scope and was able to verify that the signal makes it to the cap.  It's normally low, and when I press any key that works, it goes high then low for 100us. If I press one of the keys that doesn't work, nothing happens.  Some other pins are normally high but go low the same way when a key is pressed, other seem to be backlight and ground.
|
|
#
?
Apr 3, 2023 13:43
|
|
|
Zero VGS posted:Thank you for the insights. I have a basic multimeter with no capacitance testing, but it has continuity, and each of the capacitors is grounded on one side and not the other, so I guess that means those aren't shorted? Yep! Those should be fine. You'd also expect it tonget hot if it shorted (unless it shorted really well) but the cap is an easy thing to check. Zero VGS posted:The USB on the board is claimed to be for a touch screen; I guess if your display panel has a digitizer, the signal would go over the 40-pin EDP into that board, then out through that USB header? That would be my guess. What I can say for sure looking at the board is there's no hub or other non-power USB circuitry on it so whatever it's doing is straight pass-through. Zero VGS posted:As for the super-heating diode and inductor, do you suppose I could upgrade those with components from Mouser that run more efficiently, or at least have better heat dissipation? Maybe use components with leads, so they're not transmitting so much heat to the PCB? Can I run multiple of that SS34 diode in parallel to spread the current, without anything catastrophic happening? Does it work again once it's cooled down or does it only start up and get hot now? If it's no longer functioning I'd be worried something was damaged beyond the power supply, like the LCD itself. If it's just marginal in terms of load but everything else is working OK, I guess you could get an inductor and schottky diode with higher current ratings (not two in parallel) but it's surprising that they'd skimp on those and not the regulator itself. Like I'd want to do a lot of debugging before replacing those, like figuring out what voltages the LCD expects at what currents and whether your board is designed to provide those.
|
|
#
?
Apr 3, 2023 14:08
|
|
|
mobby_6kl posted:[*]It seems that the bottom of the caps is ground, and I can't detect anything on that side. What's the function of this setup? The cap is either part of a RC circuit to debounce the key in hardware (cap must charge/discharge via resistor limited path to change logical states), or part of an esd protection scheme (fast esd current goes through it to ground), or both The scope trace makes it look like at least part of it is for RC. The slow rise is the cap charging back up. Keyboards generally work as a switch matrix. There is a logical grid of wires and pushing a key is shorting a pair. The controller will poll with pull resistors to figure out which key it is. Foxfire_ fucked around with this message at 19:49 on Apr 3, 2023 |
|
#
?
Apr 3, 2023 19:46
|
|
|
Stack Machine posted:Does it work again once it's cooled down or does it only start up and get hot now? If it's no longer functioning I'd be worried something was damaged beyond the power supply, like the LCD itself. If it's just marginal in terms of load but everything else is working OK, I guess you could get an inductor and schottky diode with higher current ratings (not two in parallel) but it's surprising that they'd skimp on those and not the regulator itself. Like I'd want to do a lot of debugging before replacing those, like figuring out what voltages the LCD expects at what currents and whether your board is designed to provide those. It was working each time it cooled off, but the heat stress caused that R2R to become brittle and split in half when I gently placed and adjusted a heatsink. Now it is in a loop where when I plug it in, it'll turn on the blue power LED and the backlight of the display turns on for a moment, then if turns off. It's about 2 seconds on and off. So I assume that R2R inductor was pretty critical. It has continuity so the inductor coil inside it isn't totally messed up, but the core splitting seemed to ruin it. Is the current rating on Mouser the best indication of a component's heat dissipation / efficiency? I figure something might just have a higher current rating just because they let it run at a higher temperature or something. Is this a compatible diode? It has the same forward/reverse voltage as other SS34 diodes, but way higher amp rating and an integrated heatsink. It is packaged like a MOSFET: https://www.mouser.com/ProductDetail/621-SBR30A40CTFP Is this a compatible inductor? I assume any 2.2 uH is fine, and this thing looks pretty serious?: https://www.mouser.com/ProductDetail/710-7443762504022
|
|
#
?
Apr 3, 2023 21:19
|
|
|
Zero VGS posted:It was working each time it cooled off, but the heat stress caused that R2R to become brittle and split in half when I gently placed and adjusted a heatsink. Now it is in a loop where when I plug it in, it'll turn on the blue power LED and the backlight of the display turns on for a moment, then if turns off. It's about 2 seconds on and off. So I assume that R2R inductor was pretty critical. It has continuity so the inductor coil inside it isn't totally messed up, but the core splitting seemed to ruin it. Inductors in switching regulators are one of those things that have a positive feedback loop of self-destruction. The inductance decreases once the core is damaged, causing higher ripple current, causing them to get hotter, causing more damage to the core, etc, until they very quickly stop working altogether, so this tracks with my experience/expectations. Zero VGS posted:Is the current rating on Mouser the best indication of a component's heat dissipation / efficiency? I figure something might just have a higher current rating just because they let it run at a higher temperature or something. Yes. A higher-rated diode will have a lower forward drop at the same current and dissipate less heat. A higher-rated inductor will (1) have a lower ESR and dissipate less heat and (2) saturate at a higher current so, all other things being equal, it will store more energy and allow higher current output. They might also be able to dissipate more heat than the smaller ones, but ideally you wind up with a component selection that doesn't dissipate nearly as much heat for the same load. There is still one more component whose rating you can't increase as easily: the regulator IC itself (or the FET if it's external), so it may be that you replace all of these things and then that starts overheating. If you can figure out what the output voltage of this piece is, it might be easier to just desolder all the components including the IC and connect an external supply right to that output capacitor. Especially if it's supplying some common voltage like 12V. Zero VGS posted:Is this a compatible diode? It has the same forward/reverse voltage as other SS34 diodes, but way higher amp rating and an integrated heatsink. It is packaged like a MOSFET: It doesn't specify a speed, so I'd steer clear. Your switching regulator is probably running at hundreds of kHz and this thing claims in the datasheet that it's "fast" but maybe that means fast for input protection, not switching regulators. Ideally you'd have something that can switch at least as fast as the diode you're replacing. Also this one is 2 diodes in one package, so it's wasting one. Zero VGS posted:Is this a compatible inductor? I assume any 2.2 uH is fine, and this thing looks pretty serious?: Inductor looks good, but it might be a trick to get something that physically large onto the board.
|
|
#
?
Apr 3, 2023 21:55
|
|
|
Stack Machine posted:If you can figure out what the output voltage of this piece is, it might be easier to just desolder all the components including the IC and connect an external supply right to that output capacitor. Especially if it's supplying some common voltage like 12V. OK, I'm learning some stuff. I dug up the datasheet for the panel I'm using. It says for the 40-pin eDP, pins 28 to 31 are backlight ground, and 36 to 39 are backlight power (8.5-21 volts, 12v typical). Capacitor C11 (which I'm assuming is the output capacitor) has a trace which appears to run to pin 36 (you can see it in my photos), so that makes sense. I was surprised that the backlight can run on such a wide voltage range, and I thought at first maybe that's because the brightness is controlled by the voltage level you're sending into the panel, but the datasheet seems to say that the panel receives a PWM signal from this 3rd party driver board and adjusts the brightness itself from that information? https://www.dropbox.com/s/39p4eo02yamceks/B173ZAN06.8_HW0A_Ver1.0_20220107_202303136765.pdf?dl=0 I also thought this board, being specifically rated for 4K 120hz, would send an EDID signal to the PC so that it know what resolution and refresh to run at, but it seems like the EDID is actually in the laptop panel's ASIC, according to that datasheet. So, I think the entire function of this board is: 1) Pass along the eDP data signal pins (from the laptop panel) directly to the full-size DisplayPort (to the desktop PC) 2) Take 5v from USB-C, boost it to 12v, and inject that into pin 36, to drive the backlight. 3) The two Brightness +/- buttons on the PCB send a signal to the microchip nearest them, to tell it to adjust the PWM signal to the panel, so that the panel will adjust the brightness. I'm not sure if the PWM signal acts as a "pilot" (in other words, if I hook 12v to the output C11 cap like you suggested, will the panel agree to boot up and run at a fixed brightness? Or will it be pissed about not seeing a PWM signal? The datasheet details a specific "backlight sequence parameter" but I'm not sure what happens if that sequence isn't followed. Even when the board was functioning, those brightness +/- did not seem to do anything, so maybe they're miswired for this panel anyways. Amazon does sell some USB-C Power Delivery adapters which would just request 12v directly from my USC-C PD GaN wall adapter, and that certainly sounds more efficient than this board trying to boost from 12v: https://www.amazon.com/MELIFE-Type-C-Voltage-Trigger-Module/dp/B0953G14Q2 Zero VGS fucked around with this message at 00:28 on Apr 4, 2023 |
|
#
?
Apr 4, 2023 00:25
|
|
|
There's a 6-pin part and another inductor too. My guess is that's a synchronous buck providing 3.3V for the LCD VCC. The 8-pin chip over by the +/- buttons I could see the markings on clearly so I looked up and it's an 8051-compatible microcontroller. Presumably handling the PWM dimming.Stack Machine posted:The 8051 was used in, like, every damned thing from automotive ECUs to appliances for a good quarter century, and versions of it still creep into new designs. If that microcontroller needs 3V or 5V and you try to run the USB power at 12V you're in for a bad time, but that's easy enough to find out with a little prodding. Actually running the buck at 12V in may be ok, depending on whether it was designed to handle that. If you can find a part number and a datasheet that can be answered. Stack Machine fucked around with this message at 01:24 on Apr 4, 2023 |
|
#
?
Apr 4, 2023 01:19
|
|
|
That's weird to wrap my head around, I would have thought DisplayPort has it's own power sent through some of the pins, but apparently not: https://www.reddit.com/r/techsupport/comments/5k2b4g/does_displayport_carry_power/ I guess the board does need to make 3.3v and 12v in that case. Like you're saying, I might damage it trying to inject 12v myself. Seems like the safest thing then would be to replace the inductor with that one I linked (the back of the PCB is totally clean so I could stick it there), and upgrade the diode since that seemed to be tied for hottest piece on the board. Stack Machine posted:It doesn't specify a speed, so I'd steer clear. Your switching regulator is probably running at hundreds of kHz and this thing claims in the datasheet that it's "fast" but maybe that means fast for input protection, not switching regulators. Ideally you'd have something that can switch at least as fast as the diode you're replacing. Also this one is 2 diodes in one package, so it's wasting one. I don't mind wasting not using one of the diodes in the package, since that should mean double the heatsink effectiveness for the one I am using. The package looks very effective at dissipating heat (to clarify, when I think of dissipating heat I'm thinking it's ability to shed heat, but you said "a higher-rated diode will [...] dissipate less heat" which I assume you meant as heat up less under the same load as a less efficient component? Not trying to be pedantic, I'm just wondering if I'm supposed to only be using that word in an electronics context the way you're using it? The only other diode I can find that has an integrated heatsink like that, and stocked at Mouser is this one: https://www.mouser.com/ProductDetail/Vishay-Semiconductors/VS-STPS40L40CW-N3?qs=asPD7ZL2j3Wi%252BJYck45ocA%3D%3D Would that work? If not, any hints on what would? I could just buy another of these boards instead of replacing the components, but then the next board would likely still be running hot, and it takes 2+ weeks from AliExpress instead of 2 days to get components from Mouser. So it seems like the best plan is to try upgrading what I have first.
|
|
#
?
Apr 4, 2023 04:17
|
|
|
Wait why are you replacing diodes and inductors with massively larger versions? Is the theory that the power supply is grossly undersized, not malfunctioning? And that a bigger diode will fix that? (I admit I only skimmed the thread tho)
|
|
#
?
Apr 4, 2023 04:45
|
|
|
Foxfire_ posted:The cap is either part of a RC circuit to debounce the key in hardware (cap must charge/discharge via resistor limited path to change logical states), or part of an esd protection scheme (fast esd current goes through it to ground), or both I've repaired an old membrane keyboard by fixing a trace so I have a general idea how that works, but not the actual interface logic. Seems like it might be a more or less standard 24-pin setup and there's some info about it I could read when bored: https://www.instructables.com/How-to-Make-a-USB-Laptop-Keyboard-Controller/
|
|
#
?
Apr 4, 2023 10:04
|
|
|
|
| # ? May 29, 2024 18:33 |
|
|
ryanrs posted:Wait why are you replacing diodes and inductors with massively larger versions? Is the theory that the power supply is grossly undersized, not malfunctioning? And that a bigger diode will fix that? From the first time I plugged the board in it worked for 30 seconds at a time before it started screwing up and/or power cycling. Then the inductor became brittle and cracked in half. For what it's worth the seller said it's because a 17.3" screen takes too much power (literally right after he confirmed my panel was fine and shipped me the item). From that I could tell (by touching components to see what burned my fingers) it was just that final diode and inductor that were putting off so much heat. Mouser seems to go right from "tiny SMD" to "big fuckoff package" pretty quickly, and the back side of the PCB is clean, so I should be able to put both there and wire them up. But it seems like the SS34 diode has undisclosed specs like capacitance and switching speed where I'm not confident the higher-amp version will be a match.
|
|
#
?
Apr 4, 2023 12:43
|
|