|
Foxfire_ posted:ALLPCB is fine, their marketing is a little scummy though. They like to spamvertize message boards and have like 20 different storefronts that all map to the same underlying company to try to slurp up traffic. Used them once since they do aluminum core boards in prototype quantity. Yeah that's why I felt I needed to explicitly say I wasn't trying to advertise, because they offer like "$20 off your next purchase if you post a video of how great your boards are to the EEVBlog forums and mention these key marketing terms and link back here 8 times!!!"
|
#
?
Sep 19, 2018 05:27
|
|
|
|
| # ? May 28, 2024 15:26 |
|
|
ate all the Oreos posted:Yeah that's why I felt I needed to explicitly say I wasn't trying to advertise, because they offer like "$20 off your next purchase if you post a video of how great your boards are to the EEVBlog forums and mention these key marketing terms and link back here 8 times!!!" That seems unnecessary if they have sub one week turnaround to the States lol
|
|
#
?
Sep 19, 2018 05:30
|
|
|
Foxfire_ posted:ALLPCB is fine, their marketing is a little scummy though. They like to spamvertize message boards and have like 20 different storefronts that all map to the same underlying company to try to slurp up traffic. Used them once since they do aluminum core boards in prototype quantity. How do you like the aluminum core boards? I never looked into them with much detail, but I've got some high-wattage LED project ideas that might benefit from that if they're suitable.
|
|
#
?
Sep 19, 2018 06:23
|
|
|
shovelbum posted:That seems unnecessary if they have sub one week turnaround to the States lol I think their whole business model relies on someone liking them enough to do a run of 1000+ boards through them so it sorta makes sense that they'd advertise aggressively.
|
|
#
?
Sep 19, 2018 14:59
|
|
|
poeticoddity posted:How do you like the aluminum core boards? I never looked into them with much detail, but I've got some high-wattage LED project ideas that might benefit from that if they're suitable. They did what I needed to do, but there was barely anything on them. It was a LED, a protection diode, and two connectors. All the drive electronics were on a different board. It would be hard to put anything complicated on one since they only have one layer (the reverse is for clamping to a heatsink with thermal compound) It was for mounting one of these LEDs and running 1.5A through it. The manufacturer's stock board just has pads that were really annoying to get wires onto. LED manufacturer had detailed dimensions for pads, soldermask, and paste, so I was just copying that.
|
|
#
?
Sep 20, 2018 04:27
|
|
|
Sagebrush posted:istr reading somewhere that all the electrons moving through every manmade electrical circuit on Earth have a total mass of like 3kg. Similar neat factoid: When you charge a battery, it gets heavier. Same when you compress a spring.
|
|
#
?
Sep 20, 2018 16:52
|
|
|
Zuph posted:Similar neat factoid: When you charge a battery, it gets heavier. Same when you compress a spring. I mean sure but by an amount so low I don't think current human technology could measure it.
|
|
#
?
Sep 20, 2018 17:05
|
|
|
I impulse-bought this on an Amazon fire sale for about $20 earlier this week. What should I get to supplement it? General goals are 1) learning the fundamentals of electricity and circuit design (not just following kit instructions) 2) Arduino/Raspberry Pi fuckery and 3) getting my general class ham license (I have my tech but have long since forgotten everything beyond V=IR from disuse). Looks like the kit is good on sensors and widgets but has no capacitors and a smallish set of resistors, but I'm not sure what values make for a reasonable beginner set. Beyond that - transistors, diodes, more blinking lights? I already have a decent soldering iron, wire strippers, multimeter, etc. so I think I am mostly good on tools.
|
|
#
?
Sep 21, 2018 23:16
|
|
|
I could use some advice on choosing a super cheap microcontroller for a small low-budget project that hopefully results in manufacturing a few thousand units. Low cost is probably the #1 driver in this design. I looked at Octopart and the cheapest 32-bit ARM Cortex M0 (what I've used and already know) is about $0.50-60 in a 1,000 minimum order quantity. In comparison, the cheapest 8-bit micros (PIC, SiLabs, ATtiny AVR, STM-8, etc.) are about $0.35-$0.45. The minimum requirements are a built-in oscillator, a counter, 14+ GPIOs, and a few hundred bytes of RAM and flash. The rest of the design can be adjusted to match. The program it needs to run is dead simple --just basic math and driving a 4-digit 7-segment display. Am I going to give myself a headache trying to target these 8-bit chips, or are the toolchains and debug/flash tool support still considered ok? I'm not necessarily asking "what chip should I use", but "is it worth the struggle of using any non-ARM uC in tyool 2018?" Thanks 
|
|
#
?
Sep 22, 2018 00:10
|
|
|
Is the per unit cost saving worth the time it will take to learn the new tools and architecture and make any tool purchases? I kind of expect that unless you're using something obscure and barely-supported that the toolchain won't be bad enough to really factor into the decision, compared to those factors. Even if the tools are great there's still a lot to learn after all. I think the Microchip ICD series of debuggers are pretty excellent pieces of hardware and I think the software is fine (though the old MPLAB had a better interface for some of the debugging tasks and a nicer code disassembly listing) and I've never started a project and thought "god drat these tools are a hassle to have to use." Microchip also has some of the nicer datasheets among micros that I've ever seen. But then you've still gotta pay for that debugger and learn a new IDE and read the datasheet to learn the peripherals and how to set up the oscillator and such. Is that worth the savings?
|
|
#
?
Sep 22, 2018 00:34
|
|
|
Yeah, I agree with the above. Atmel have amazing datasheets and you'll smash the design out in a few weeks using some sort of ATTiny. Or you can use something poo poo like a GD, which is a Chinese company making clones of STmicros except with more silicon bugs. A couple of thousand is still pretty low volume so it depends on the skills and experience of your dev team and how much time you have.
|
|
#
?
Sep 22, 2018 11:45
|
|
|
Cool, thanks both of you! I’m doing the whole project solo so it’s really about personal struggle and pain.
|
|
#
?
Sep 22, 2018 18:38
|
|
|
ante posted:99% ISO alcohol and an old toothbrush longview posted:Ultrasonic cleaners work pretty well for cleaning PCBs too, I run mine with 99% iso for PCBs. Late to the party but... how about an old ultrasonic toothbrush? ps. Nail polish remover is not a good substitute for pure ISO alcohol. 
Normal Barbarian fucked around with this message at 00:55 on Sep 24, 2018 |
|
#
?
Sep 24, 2018 00:05
|
|
|
Yeah there isn't a lot on a circuit board other than the metal parts that acetone probably wouldn't dissolve.
|
|
#
?
Sep 24, 2018 00:15
|
|
|
Fortunately my (sister's) nail polish remover was over 20 years old by the time I got to it.  --- Speaking of old things that don't do anything, my robot is finally taking the leap from utterly-naive retail therapy to mostly-naive almost-reality, and I have some questions. 1) My 8 IR distance sensors draw 30mA typical and 50mA max, but they pulse in excess of 300mA. Do Pololu's carrier boards compensate for that, or do I need to do more?  2) I have a handful of (~20mA?) RGB LEDs split between an nrf52 Bluefruit feather and a Pi Zero. The feather can source/sink an absolute max of 10mA per pin (5mA recommended) and an absolute max of 30mA for the entire package. The Pi seems to have similar limits. Toshiba's magical DMOS chips seem like a good solution, but I'm intimidated by this sort of thing:  I don't know how to choose any of those components. How do I go about determining what I need? Do I really need those ESD diodes? Should I give up on the Toshiba and use individual MOSFETs? 3) Finally, I think I want a fuse of some sort, probably. Between two steering servos and two ESCs driving two RC rock crawler motors, is there any way estimate current, or do I have to wait until everything is assembled and running and stalling?
|
|
#
?
Sep 24, 2018 01:10
|
|
|
For #1, what do you mean by compensate? There doesn't seem to be any power supply stuff other than filtering on that board so I can't imagine the board cares unless you're drawing enough current to burn traces or something. The only places the supply pin connects to is the VCC and anode of the sensor module, a pull-up on a logic signal, and the aforementioned filtering. For #2, I can't imagine why you'd need those protection diodes if all you're doing is using them for is to drive LEDs from a low-current rated logic signal. For #3, I don't know about those parts in particular, but when I'm trying to estimate current for parts that have datasheets I use the maximum current draws listed (if any) and add them all together and add a margin of error (like multiply the total by 1.1 or 1.2) and use that figure. If any of your parts don't have good datasheets, then you may have to do some testing of your own. Hook up the parts individually and see if you can figure out a peak current for each one, add them all together, and then multiply the total by a margin, maybe a bigger one if you're not so confident. edit: essentially in engineering you want to add a bunch of margin whenever possible to make sure your estimates are never on the bad side of being wrong BattleMaster fucked around with this message at 01:32 on Sep 24, 2018 |
|
#
?
Sep 24, 2018 01:27
|
|
|
BattleMaster posted:For #1, what do you mean by compensate? There doesn't seem to be any power supply stuff other than filtering on that board so I can't imagine the board cares unless you're drawing enough current to burn traces or something. The only places the supply pin connects to is the VCC and anode of the sensor module, a pull-up on a logic signal, and the aforementioned filtering. 1: I was scared by the 300mA number, which seems large, and that 8 sensors pulsing (spiking?) 300mA might mess with the other sensors/bluetooths/Pi's somehow. I've been reading about inrush current, and I probably did some conflating. 2: Excellent. 3: Sounds good, I'm down for some empiricism-ing. From what little I've read, stalled motors and sensors can draw hella current, and fuses/breakers rated for hella current seem large and expensive. That's probably something I'll just have to live with. Normal Barbarian fucked around with this message at 03:09 on Sep 24, 2018 |
|
#
?
Sep 24, 2018 02:57
|
|
|
If it just takes a big gulp of current when first powered up, consider doing some sequencing so they don't all come on at the same time or have something set up so the electronics that may be more sensitive to power rail dips are held in reset until things have a chance to stabilize.
|
|
#
?
Sep 24, 2018 03:06
|
|
|
meatpotato posted:I could use some advice on choosing a super cheap microcontroller for a small low-budget project that hopefully results in manufacturing a few thousand units. Low cost is probably the #1 driver in this design. Jay Carlson has probably the best overview of cheap microcontrollers that has ever existed: https://jaycarlson.net/microcontrollers/ I'm partial to the Atmel tools myself, and you've got a lot of options in that area. If you're really opinionated about open-source and/or Linux-compatible toolchains, ARM is probably the winner.
|
|
#
?
Sep 24, 2018 14:14
|
|
|
Zuph posted:Jay Carlson has probably the best overview of cheap microcontrollers that has ever existed: https://jaycarlson.net/microcontrollers/ The only problem I have with Atmel is that they keep releasing new programming interfaces that nothing but their internal tools (which only run on Windows) support. I bought their fancy programmer and a bunch of attiny 1616's without realizing this and now I have no easy way to program them 
|
|
#
?
Sep 24, 2018 14:52
|
|
|
Zuph posted:Jay Carlson has probably the best overview of cheap microcontrollers that has ever existed: https://jaycarlson.net/microcontrollers/ That overview is just insanely good. Thank you.
|
|
#
?
Sep 25, 2018 04:09
|
|
|
I'm going to be using a Pokeys57U as the core of my custom KSP controller build. It has everything I need: 55 inputs configurable as digital I/O OR keyboard emulation, 7 analog inputs designed for 3.3v joystick pots, etc. Very handy, and lets me avoid using yet another Arduino (plus, I am bad at programming). Also can run basic programs to drive the digital outputs (in an easy block-based graphical language for my monkey brain), and the big one to me is that it's reusable and versatile. Can also be set up for matrices, and can run those 2 line LCD displays you see on every hacker project nowadays. Can set the inputs and outputs to be high or low, 5V or 3.3V, per pin. However! I would like to run some individual indicator LEDs with it on my spare pins, as the controller only uses 28 or so of the buttons (it'll be emulating a keyboard, with joystick axis, for maximum cross-game compatibility). Standard red or green things, I know how to wire a basic current controlled LED circuit, yadda yadda. But I've never done it with logic level voltage and currents before, on individual LEDs. The Pokeys57 has a limit of 5V 4mA, or 3.3V 12mA on digital outputs. My kneejerk reaction is to find a handful of logic-level MOSFETs on ebay or digikey or whatever, and use those (but with each individual LED instead of a string like I've done before). Does that sound right? I have a handful of IRL540 FETs hanging around, but don't really want to have to heatsink them. I guess I could use the alu case of the controller or something. Was planning on using those prepackaged indicator LEDs set up for 5V at 20mA or so each (from my foggy history with them). That crazy? Queen Combat fucked around with this message at 06:35 on Sep 25, 2018 |
|
#
?
Sep 25, 2018 06:31
|
|
|
Yeah, sounds like you're expecting to use a crazy amount of power for LEDs. First step, find the LEDs you want to use so you can design around them. Next, I kinda expect that a shift register or 5 will be fine to buffer, and also, you know, shift values to simplify wiring and not waste pins needlessly. The 74HC595 or something
|
|
#
?
Sep 25, 2018 06:47
|
|
|
While it does support shift registers, it's sorta-proprietary in its implementation (they push an addon board for easy integration, though you can command-line program your own if you want, from what I gather). However, I'm only doing 7 indicators max, and they're actually 6mA 5V indicator LED assemblies that I've had kicking around for awhile, so I'm not short on pins with 30 to spare. The 20mA figure I was pulling from an old aquarium lighting project I'd done a few years ago, my bad.
|
|
#
?
Sep 25, 2018 07:10
|
|
|
Bunch of 2N3904 BJTs?
|
|
#
?
Sep 25, 2018 07:17
|
|
|
Hot drat that's a better, cheaper idea and I think I've still got most of a 100-strip of them in a box somewhere. They're fast enough to do some PWM animations, too. Thanks! Don't know why I was overthinking it so much. Queen Combat fucked around with this message at 07:30 on Sep 25, 2018 |
|
#
?
Sep 25, 2018 07:26
|
|
|
I went with 2N7000 N channel FETs on my LED driver board.
|
|
#
?
Sep 25, 2018 14:05
|
|
|
Mr. Powers posted:I went with 2N7000 N channel FETs on my LED driver board. Those have like 5 ohms Rds(on), It'll do the job if they're what you have laying around but imo if you have any more modern option take it instead
|
|
#
?
Sep 25, 2018 14:53
|
|
|
Does anyone keep a database of the release dates of parts, especially older 20th century standbys like the 2N3904? I'm kind of curious about how old some of the part designs are.
|
|
#
?
Sep 26, 2018 00:09
|
|
|
BattleMaster posted:Does anyone keep a database of the release dates of parts, especially older 20th century standbys like the 2N3904? I'm kind of curious about how old some of the part designs are. It's amazing how much old stuff gets used in hobbyist designs out of seemingly sheer inertia and through hole availability. There's a world of components out there that a lot of people probably are intimidated by even though they can be drag soldered onto a breakout board by a monkey.
|
|
#
?
Sep 26, 2018 00:39
|
|
|
shovelbum posted:It's amazing how much old stuff gets used in hobbyist designs out of seemingly sheer inertia and through hole availability. There's a world of components out there that a lot of people probably are intimidated by even though they can be drag soldered onto a breakout board by a monkey. The worst one is the 741 op-amp when even other really old (but not as old) op-amps like the LM358 are a ton better.
|
|
#
?
Sep 26, 2018 01:35
|
|
|
Finding new parts is a pain so people get set in their ways and pick their favourites. I solve this problem by never remembering part names so I am forced to go and find what I want on digikey.
|
|
#
?
Sep 26, 2018 02:00
|
|
|
shovelbum posted:It's amazing how much old stuff gets used in hobbyist designs out of seemingly sheer inertia and through hole availability. There's a world of components out there that a lot of people probably are intimidated by even though they can be drag soldered onto a breakout board by a monkey. I picked the 2N7000 because I am not good at semiconductor design and saw some existing designs using it. I never thought to check RDS(on). The surface mount parts we use at work are always way down in the milliohms, so it never even occurred to me that there might be old parts that aren't actually very good when it comes to FETs. E: I did some soldering at work today and this is my finest work yet (finest as in smallest, not as in best/cleanest). I removed a W.FL connector and rotated in 180 degrees, and i'm pretty sure it would have works if the rotation were not the issue:  For comparison, this one is untouched: 
carticket fucked around with this message at 02:09 on Sep 26, 2018 |
|
#
?
Sep 26, 2018 02:04
|
|
|
Mr. Powers posted:I picked the 2N7000 because I am not good at semiconductor design and saw some existing designs using it. I never thought to check RDS(on). The surface mount parts we use at work are always way down in the milliohms, so it never even occurred to me that there might be old parts that aren't actually very good when it comes to FETs. W.FL is so cool, what a nice solution to the massiveness of SMA or whatnot. Yeah, I agree with you there, you could probably go down a rabbit hole of JUST looking at FETs for as long as it would take to have a working prototype.
|
|
#
?
Sep 26, 2018 04:34
|
|
|
Also, as a note, I'm pushing this at work, but I want it to spread beyond. W.fl is pronounced "waffle". Keep this spreading across the electronics industries.
|
|
#
?
Sep 26, 2018 17:11
|
|
|
Splode posted:Finding new parts is a pain so people get set in their ways and pick their favourites. I solve this problem by never remembering part names so I am forced to go and find what I want on digikey. I mean I think it's also just because 90% of common hobbyist designs trace their origins to stuff from like, 1960 - 1990, so everything online is one of those originals or a copy of a copy of a copy. So you sit down to build your first thingy and it uses them, so you buy 1000 2N3904's or whatever since they're so cheap, and now for every project moving forward it's easier to use one from the pile than order new ones, so now even your new designs that you publish use them, etc. I did that a bit at first but quickly became obsessed with min/maxing parts so now I have all sorts of weird poo poo
|
|
#
?
Sep 26, 2018 17:17
|
|
|
Basic question - I have a 10mm flickering LED and a CR2032 battery I'm using for a halloween prop. What kind of lifespan should I expect out of the battery? Here's the usage: https://www.youtube.com/watch?v=O1h3bvH6jnE
|
|
#
?
Sep 26, 2018 17:41
|
|
|
Couple weeks?
|
|
#
?
Sep 26, 2018 17:49
|
|
|
Awesome, thanks. I just wanted to make sure it wasn't a few hours or anything.
|
|
#
?
Sep 26, 2018 17:50
|
|
|
|
| # ? May 28, 2024 15:26 |
|
|
If you have a multimeter you can check the current it uses (unless the current is too low to be measured by your multimeter, cheap ones might not be able to read it) and then calculate about how long the battery will last using the milliamp-hours rating of the battery. A quick google says it's 235mAh for CR2032's made by Energizer, which is probably not gonna vary much between brands.
|
|
#
?
Sep 26, 2018 18:32
|
|