|
Finished my first major project solo. Going to send it out in the next few days after obsessing over everywhere I could have messed up. 50/50 chance I unknowingly short something when the solder flows! The biggest planning mistake was the layout of the LED shift register drivers. I should have just thrown everything on the right rather than spreading it out. Had I been able to just route them left to right a lot of real estate could have been saved. Second biggest planning mistake was avoiding placing anything but the LEDs on the front. Even knowing I would be soldering components on both board I crammed everything in the back. The front is just 1206 LEDs though, so small blessings. Third biggest was sticking to mostly 1206 components for assembly ease even though I'm pretty confident working one step down at 0805. The biggest routing mistake was holding off too long on routing all the clock independent digital lines which wound up redoing the bottom center and moving the microcontroller and more knock on results. The one question that this project raised that I never thought to ask was whether to route oscillator traces as matched length differential pairs or just as short as possible. I threw a little kink in the 32khz crystal to make up for a 1mm difference in routing and don't know if I really need to do that. Probable just makes it worse since I didn't find any good google results before hand.  https://i.imgur.com/BbSXOJy.png  https://i.imgur.com/sPrug07.png  https://i.imgur.com/y45Fsha.png M_Gargantua fucked around with this message at 07:24 on Mar 16, 2019 |
#
¿
Mar 16, 2019 07:16
|
|
|
|
| # ¿ Apr 26, 2024 00:25 |
|
|
TI also has 8 and 16 bit shift registers that have a grouped PWM input. The ones I’ve used have one PWM input per bank of 8. Easy to do brightness control and dimming. And in rare the case that you absolutely must have different duty cycles on each individual output the serial rate and PWM response time mean you could keep rotating through different outputs faster than the eye can tell.
|
|
#
¿
Mar 25, 2019 20:59
|
|
|
You refered to them as button data and led data, but those seem like regular buttons. I doubt they have any internal circuitry that handles "data". A shift register will work great for your purposes. You could go with one 16 bit shift register for the LEDs, and either a 24 bit or a 16 + an 8 bit for the button multiplexing where you have them all matrixed to one Input pin while continually sweeping which button/potentiometer is excited and read at any given moment. With a good parts selection the sweep time will be able to capture each input multiple times before the hardware bounce of the button presses even settles, so you won't notice any input lag on that end.
|
|
#
¿
Mar 26, 2019 05:25
|
|
|
How much power do you need to transfer is the big limitation here. If you want to fast charge it during a day of hard use the wireless idea isn't going to work. But if you only need it to top up and maintain over 24hrs you can build a simple inefficient one and that'll be fine.
|
|
#
¿
Apr 5, 2021 20:17
|
|
|
The footprint editor will handle all the layers for you if you don’t fight it. Just don’t do something silly like turn off solder mask. For that footprint one thing I would say is to not have any silk screen directly abut a pad like that. Best practice is to give it the same clearance you would trades pads. Or like 0.15mm. Not that it’s going to bite you on something big like a switch, but silk misalignment can cause an issue if it prints on a pad as you get into smaller stuff.
|
|
#
¿
Apr 12, 2021 02:18
|
|
|
Too add a little bit about the layers Courtyard layer is the area where it will cause mechanical issues with assembly with nearby parts, and must be a single continuous boundary. Fabrication layer should be matching the mechanical drawing Silk should be placed so that there is nothing under anything the fab layer covers, as silkscreen under a part is only useful on initial hand assembly, and screw you for future troubleshooting Copper is the manufacturer recommended solder pad for the package and Mask will be the copper + whatever the manufactures recommended solder mask clearance is for that part and package. If you're playing nice you can add a note in either the commends layer or the drawings layer for what the stencil thickness would be for that footprint mask that way you can know to change it when you have a few parts that the manufacturer specified a thicker/thinner stencil then the bulk. Pro level is you put the part find number in the drawings layer and when you make a board you drag all the drawing find numbers over to a BOM in your drawings layer, and hence get an engineering drawing out of the deal.
|
|
#
¿
Apr 12, 2021 02:26
|
|
|
How would you make an automatic shower cleaner that doesn't just get water everywhere all streaky and gross. How would you get it to actually scrub things? So many unanswered questions.
|
|
#
¿
Apr 19, 2021 02:33
|
|
|
So the scrubbing bubbles just sprayed soap everywhere? That doesn't clean poo poo. Please take a brush to your showers now and then goons, like maybe once a month? 4 times a year, I can compromise?
|
|
#
¿
Apr 19, 2021 05:07
|
|
|
Also sure I'll make one if people are dumb enough to buy it cause god what a garbage product. This is not difficult engineering at all.
|
|
#
¿
Apr 19, 2021 05:08
|
|
|
Could you be more specific on what kind of switching you want? Or why? Contacts inherently act like switches in a way, and what your describing sounds like depth keying, where certain pins extend further so they make contact first and act as your switches/grounds/shields whatever.
|
|
#
¿
Apr 23, 2021 22:44
|
|
|
One weird trick(tm) that I’ve seen done once is taking a cheap automotive direct injection fuel pressure sensor and potting it with something that transfers the pressure but blocks the liquid. I think it was like a 3d printed cup and some sort of jello’y urethane diaphragm. Some sort of tiny pinhole diffuser in the 3d printed piece so that the liquid couldn’t direct impinge on the diaphragm either. No idea the exact specifics or how long it lasted once installed but I did at least see it work. Or just use the sensor directly on a bench and see how long it lasts? They’re like $30
|
|
#
¿
Apr 27, 2021 16:58
|
|
|
Disregard that I’m an idiot This is the perfect use case for just making a precision inline piece of tube in your gas system and getting pressure by measuring the expansion of it via strain gauge. Add in a temp compensation and you’ve got a highly accurate system for next to nothing besides hours spent machining
|
|
#
¿
Apr 27, 2021 17:06
|
|
|
Ambrose Burnside posted:i am wondering if I could use strain gauge on some existing part of the gun to achieve a similar end Absolutely yes. Find one section that has a gas volume at the pressure you want. Because you didn’t engineer the part you’ll have to calibrate it manually and extrapolate. Either two or four strain gauge rosettes with temperature compensation pads, and a thermocouple. The temp compensation pads correct for the difference in resistance in the wires as the surface changes temp. The thermocouple is so you can compensate for the change in temperature of the part itself. Then hook up your system to a stationary pressure source with a good gauge and slowly raise the pressure and log what the strain gauge sees. Do it at a couple different temperatures. Should get you a nice 2D map of P = f(T, ε) The rosettes are really cheap, it’ll be finding a low cost strain gauge amp and TC amp that might be difficult, but it’s not like you need lab grade instruments. Slapping a few off the shelf ICs on a PCB to amplify the signal to a level to be read by your MCU might get you +/- 5% with poor stability but since you’re calibrating it against actual conditions it’ll be more than accurate enough for what you need.
|
|
#
¿
Apr 28, 2021 22:52
|
|
|
lunael1982 posted:Hi! Easiest answer is to throw on some EEVblog/Great Scott/ScanLime/Carl Brgeja/Electroboom/etc youtube in your background and you'll pick up some stuff by osmosis in the 1000 hours of content. There are a ton of books and stuff but for general learning I always recommend youtube because moving pictures are less boring for the beginner. I will gladly read research paper on RF antenna design but no way i'd recommend it to someone who wants surface level info. Sounds like general people making things youtube might also be more relevant that electronics directly. This Old Tony, AvE, Make it Extreme, etc. Lots of them out there that are wonderful if you avoid some of the dumber channels
|
|
#
¿
May 4, 2021 22:33
|
|
|
Get a nice microphone + use OBS to record the audio from a VoIP call
|
|
#
¿
May 7, 2021 08:38
|
|
|
Scanlime has a far better calming narration voice than Big Clive. Fight me  Also very very long technical livestreams that include their cat. e; This was the video four years ago that introduced me to the channel https://www.youtube.com/watch?v=TeCQatNcF20 but all the extended tech hacking live streams are in the scanlime-in-progress branch channel M_Gargantua fucked around with this message at 17:45 on May 9, 2021 |
|
#
¿
May 9, 2021 17:42
|
|
|
If you are flat mounting the film there’s no reason for anything to be rotating like that. Drum scanners work by spinning the film over the scan head so you cover all the surface, but if the film is flat there is no reason to spin the scan head instead. Just sweep it linearly across the XY like all the other modern CNC devices with our modern precision position encoders. As for the flattening can you just use a very specialized vacuum table to hold it perfectly flat?
|
|
#
¿
May 11, 2021 12:11
|
|
|
They’re mostly just all-in-one IC amplifiers these days. Things have been miniaturized and combined and mass produced. Hell they probably make a SOT-23-6 or even smaller one that would be overkill now. You’re thinking too big. The altoid can is a three day battery pack attached to a rice grain
|
|
#
¿
May 17, 2021 10:49
|
|
|
Just did some checking and they have <$2 ones with AGC and all now. But if you want AGC you need a “large” DFN package for example. Technically I’m sure a microphone amp could consume 2W? A 1S altoid can sized lithium would be like 6000mAh so a rather excessive 8 hours of continuous use at peak power.
|
|
#
¿
May 17, 2021 13:58
|
|
|
The MAX9814ETD was the first result of my search and that’s good enough for what I was looking for
|
|
#
¿
May 17, 2021 14:39
|
|
|
Sure you could make a far inferior opamp & protoboard version, but why? Things being through-hole protoboard doesn’t really make it any easier of a DIY project like it would be 20 years ago
|
|
#
¿
May 17, 2021 15:34
|
|
|
petit choux posted:Why the gently caress would it be "far inferior"? Cost. Sensitivity. Power consumption. Part count. With what I spent 30 seconds for on digikey you’ve got a 7 component + power source solution to the stated problem and fits on a PCB the size of the 3.5mm jack itself. Small enough that you can just put the electret mic right on the board if you’re not going to be wearing it. 1S lipo is an ideal power source and if you for some reason want a built in charging Circuit that’s only three components more. If you want a Headphone amp and not a microphone amp you’re up to 18 components and higher power consumption but it’s still tiny. You don’t need a “kit” for an IC and a few capacitors M_Gargantua fucked around with this message at 17:56 on May 17, 2021 |
|
#
¿
May 17, 2021 17:53
|
|
|
Forseti posted:Boy am I going to regret wading into this one, but IMO: This post is too many words to say the obvious! Inefficiency! Inefficiency!
|
|
#
¿
May 17, 2021 21:52
|
|
|
KnifeWrench posted:Quick question that I'm having a hard time searching: I have a system that's charging LiFePO4 batteries with 54V @ 25A using spring contacts. If contact is broken unexpectedly, my intuition is that intermittent or poor contact during the transition (i.e. bouncing or nontrivial impedance) is bad news that could cause arcing or otherwise deteriorate the system. Nothing catastrophic, but possibly lifetime shortening. Yeah for a battery pack like that, standard design practice would be to include a fuse + accumulator isolation relay (AIR) that cuts power through the spring contacts before any motion could potentially break the connection. You don't want to be accidentally breaking and making that sort of contact under those loads. Hopefully there is a pretty strong lever arm clamping the pack to the charger? e; AIR is just a use case term, not a type of relay. Its just a large relay used to interrupt flow to and from any major power accumulator such as a battery or capacitor.
|
|
#
¿
May 18, 2021 18:52
|
|
|
Shame Boy posted:
Rivets! Far less obvious.
|
|
#
¿
May 20, 2021 23:04
|
|
|
Rather than make the flange bigger you can also just use a smaller 4:3 screen and a cheap plastic lens to stretch the image a bit to fill the full circle, do it right and you'll only lose some clarity at the edge cardinal directions.
|
|
#
¿
May 21, 2021 21:55
|
|
|
I have seen some similar circuits with feedback mic's/vibrometers downstream like that, but those feed into tiny DSP's to adjust the phase shift and balance of the sound. I'm sure you could do it full analog but generally the systems are meant to be small which is why you get the DSP IC.
|
|
#
¿
May 26, 2021 18:09
|
|
|
Using a 2.5mm barrel for AC power makes me sad. We're getting so close to standardizing 2.5mm as the 12V plug. (Just like C7's and C13's are for 120V((gently caress you C5's!  )))Using a rectifier on the board that does the amplification is bad engineering practice and bad for sound quality and hence makes me sad. Having to provide non-standard AC voltage to your product makes me sad. Almost everything in the audio world makes me either sad, or in cases of poor engineering practice that is merely a carryover from yee-olden days that continues to get used, just angry I'm going to design that AGC condenser mic preamp I brought up a page ago for you all just so I'm not just here with nothing but twitching veins and complaints. e; Sagebrush posted:I told you at the beginning to find a wall wart at Goodwill or something and cut it open to get the transformer and you said no. The board shows an AC-GND-AC which thanks to the poor documentation of chineseium could go either way. I can't visibly see why it wouldn't be 12-15 rail-to-rail unless by GND they mean neutral and the board is ungrounded, which is equally likely. M_Gargantua fucked around with this message at 15:04 on May 28, 2021 |
|
#
¿
May 28, 2021 15:01
|
|
|
Shame Boy posted:Is there a good standard connector for isolated low voltage + / - with grounded shield? The only thing I can think of are XLR or these cheap lovely connectors that are all over Amazon: You'll probably be good with like a DIN connector? The product answer is MIL-DTL-38999 series 3 connectors. But "standard" is rarely ever "economical". They are unarguably best though, Brought to you by Amphenol, Glenair, & TE each sells cross-compatible parts. Tooling will run you a cool $200 for the crimp and positioner and the connectors themselves get real expensive if you start adding in options like backshells and strain reliefs. However, if you get a crimp & positioner they're also the same type of tooling for the beautiful (unshielded) plastic equivalents to 38999's everybody has, as well as things like Deutsch's DTM/DT/DTP series which are another highly recommended system. Just remember to be very specific with your part numbers and consult the multi-hundred page catalogs of different arrangements and inserts.
|
|
#
¿
Jun 2, 2021 02:12
|
|
|
petit choux posted:Cool, thank you. Now here's my next dumb questions, and sorry if we've talked about this before but I couldn't find it a few weeks ago: KiCad and PCBway are the most reliable combo i've found. I'm not a big fan of Eagle on the software side. OSHpark is alright if you just need a temporary test board, but they have a severe lack of options compared to a big house. Advanced Circuits has consistantly given me garbage quality boards whenever i've used their cheaper tier (which is still quite a bit more expensive than PCBway even including shipping)
|
|
#
¿
Jun 25, 2021 13:50
|
|
|
Arduino: pretty mediocre really
|
|
#
¿
Mar 28, 2022 03:07
|
|
|
Chassis should be grounded to the earth ground of the system u less you have good engineering reasons for doing anything else.
|
|
#
¿
Mar 28, 2022 12:35
|
|
|
Slanderer posted:Does anyone know what the easier way to quickly cut a few D-shaped holes in enclosure panels (for mounting potentiomers). Are there affordable off the shelf punches, or is milling the best bet? Maybe just design and order a new enclosure front from SendCutSend?
|
|
#
¿
Apr 27, 2022 04:36
|
|
|
|
| # ¿ Apr 26, 2024 00:25 |
|
|
The vias near pins 12 and 14 also look like they probably violate minimum spacing to the other surrounding copper. You can always add a wiggle or a narrow but to a trace at spots like that.
|
|
#
¿
May 2, 2022 00:58
|
|