|
Those are what was I was thinking of. I don't see why you'd need an LDO at all with one of those switching solutions unless you want the LDO for a normal reason like needing a second lower voltage that is low power. The LTC3331 looks pretty good. I think babyeatingpsychopath was referring to a different architecture.
|
#
?
Apr 7, 2020 20:47
|
|
|
|
| # ? Apr 16, 2024 20:48 |
|
|
I've used the LTC3108 and it's for the opposite problem that you have - the designed input current is high while the input voltage is very low. That's what you'd expect from a thermoelectric generator or a small solar cell, which is what it's made for. I have no experience with them but I'd imagine that the piezoelectric-focused ICs would be more towards the higher voltage, low current end of the spectrum.
|
|
#
?
Apr 7, 2020 21:03
|
|
|
taqueso posted:Those are what was I was thinking of. I don't see why you'd need an LDO at all with one of those switching solutions unless you want the LDO for a normal reason like needing a second lower voltage that is low power. The LTC3331 looks pretty good. I was honestly thinking he was going to be getting 3.5V max and needing stable 3.3V; the fact that he can get somewhat arbitrary voltage and current out of this thing is pretty awesome. Ambrose Burnside posted:- these long cell linkages i'm prototyping are an inefficient use of chain length. I could probably make them half as long as it stands, but I think I wanna tweak the design to use wide, 'shallow' linkages w maybe two terminal contacts (or one 'circuit' link, one floating 'dummy' structural link) per side, so think square/rectangular cells with two linkage-terminal loops on two opposed cell sides. More like the "improved pulvermacher chain" design that his original patent developed into, now that I think of it: Have you considered putting tiny rubber bands on the spots where the links connect to preload the contact points a little bit?
|
|
#
?
Apr 8, 2020 15:54
|
|
|
babyeatingpsychopath posted:I was honestly thinking he was going to be getting 3.5V max and needing stable 3.3V; the fact that he can get somewhat arbitrary voltage and current out of this thing is pretty awesome. Yeah, apparently the real primary users of the original 1800s product were early electrical experimenters, who saw its utility as a perfectly-modular + infinitely-scalable DC power supply. 100+ link batteries were regularly assembled and used, i’m taking a less spectacular approach And yeah, some sort of preloading is def needed if i’m not starting w close-conforming material profiles. I‘ve been trying to find a way to do it with the electrode wire itself, something like winding one electrode end into a flat coil spring and pressing it inside the rigid loop of the opposed electrode, and constraining everything with twine/elastics/etc like i have so far
|
|
#
?
Apr 8, 2020 16:44
|
|
|
Ambrose Burnside posted:Yeah, apparently the real primary users of the original 1800s product were early electrical experimenters, who saw its utility as a perfectly-modular + infinitely-scalable DC power supply. 100+ link batteries were regularly assembled and used, i’m taking a less spectacular approach Bent-wire joints were all the rage around this time. https://patents.google.com/patent/US732462 Note that this patent is expired, so go hog wild!
|
|
#
?
Apr 8, 2020 17:09
|
|
|
There was a time you could patent bent wire? What was that even for? Clothes fasteners?
|
|
#
?
Apr 8, 2020 17:14
|
|
|
kid sinister posted:There was a time you could patent bent wire? What was that even for? Clothes fasteners? You can still patent bent wire if you do it right. You just need to combine it with something else. For example, I was told by a SRAM EE that Shimano has a patent on parallel stacks of PCBs on a bicycle so sram used perpendicular PCBs in their e-shifter design.
|
|
#
?
Apr 8, 2020 17:18
|
|
|
kid sinister posted:There was a time you could patent bent wire? What was that even for? Clothes fasteners? If Ambrose comes up with a novel way of connecting pulvermacher chains together in a mechanically and electrically sound manner, he could still patent it. It's even legal to say "I derived my electrical chain socket joint from existing patent XXX now in the public domain" and roll with it. That's if he thinks this is commercially viable and wants to prevent anyone from stealing the idea.
|
|
#
?
Apr 8, 2020 17:42
|
|
|
He didn't point at each of us and say "patent pending" at each person, so he's pretty much screwed.
|
|
#
?
Apr 8, 2020 18:25
|
|
|
What's the scoop on sizing up a replacement barrel adapter? I have an old mp3 player unit I want to try and revive, but don't have a matching power supply for it on hand. What I do know is that I'm after a 5v supply, up to 2 Amps, with positive in the center. My dodgy calipers tell me the diameter of the hole on the mp3 player is about 4.25mm across, and about 7.75mm deep. I can't get the calipers around the pin in the center, but it looks to be about 1mm wide if I eyeball it against a 2mm wide drillbit. This might be a case for one of those adjustable voltage wall warts with a collection of swappable heads, and take a chance?
|
|
#
?
Apr 8, 2020 19:53
|
|
|
babyeatingpsychopath posted:If Ambrose comes up with a novel way of connecting pulvermacher chains together in a mechanically and electrically sound manner, he could still patent it. It's even legal to say "I derived my electrical chain socket joint from existing patent XXX now in the public domain" and roll with it. nothin groundbreaking so far, but it's fun experimenting w different approaches. turns out making springs with hammer + anvil and literal garbage from behind the workshop bench is not the pareto-optimal way to approach this. copper isn't a great spring material so i have to work-harden the poo poo out of it to get useful elastic deformation instead of just plastic deformation. for most of these i'm starting w round wire and then hammering it into a flat rectangular profile and rolling that up like a clock mainspring, the idea being to max out the potential contact area w the similarly-profiled magnesium ribbon; my other thought is to use plain round wire springs but use them in pairs inside each Mg terminal loop  i suspect it'll be easiest to design the terminal-preloading component as its own double-ended coupling linkage used to join all cells to one another. I need a semipermanent way to connect cells while letting me modify the battery down the road, right now the outer sheathing prevents opening up the cell terminal loops post-assembly so i can't "mix and match" cells without having to repair em afterwards. also wish I had some bronze wire, it's what they use for leaf spring-type battery contacts b/c it's got the mechanical properties you want plus decently-low resistance. nickel silver too, esp b/c its oxide is still highly-conductive. for all i know that'd screw with the galvanic chemistry at work here, tho Ambrose Burnside fucked around with this message at 21:15 on Apr 8, 2020 |
|
#
?
Apr 8, 2020 20:36
|
|
|
Shine on you sweaty diamond
|
|
#
?
Apr 8, 2020 20:47
|
|
|
RedChesterfield posted:What's the scoop on sizing up a replacement barrel adapter? I have an old mp3 player unit I want to try and revive, but don't have a matching power supply for it on hand. Yeah just go with one of those universal ones with 10-15 different tips and there's a 99% chance at least one of em' will fit. There's a few rare barrel plugs that are specifically designed to be proprietary and incompatible but this doesn't sound like one of those.
|
|
#
?
Apr 8, 2020 21:02
|
|
|
babyeatingpsychopath posted:That's if he thinks this is commercially viable and wants to prevent anyone from stealing the idea. You mean that's if he thinks this is commercially viable and wants to have the opportunity to go to court against anyone stealing the idea. Having a patent doesn't do poo poo if you don't actively enforce it and have the legal expertise and resources to do so.
|
|
#
?
Apr 9, 2020 00:17
|
|
|
LDO has become a generic term for regulator even though most regulators arn’t used in actual low dropout applications.
|
|
#
?
Apr 9, 2020 01:52
|
|
|
Hey bros I was thinking about the short HDMI / video/ USB etc signal traces and strict requirements. Think this is related to why eyes/ears/nose are very close to the brain on nearly every vertebrate?
|
|
#
?
Apr 9, 2020 02:07
|
|
|
That's part of it - Shorter distance to travel, faster the signal gets there. That would be the case if you put a DDR4 chip on there, for example. For USB and HDMI, though, the signals are also going along these really long cables. The reason we were recommending short lengths is that the differential traces are required to be really really closely matched to each other, both in length, and in impedance. The longer your runs are on the PCB, the better chance of that getting hosed up. Cheap boardhouses won't even properly control your impedance.
|
|
#
?
Apr 9, 2020 02:31
|
|
|
Dominoes posted:Hey bros I was thinking about the short HDMI / video/ USB etc signal traces and strict requirements. Think this is related to why eyes/ears/nose are very close to the brain on nearly every vertebrate? There are a number of reasons an intelligent designer (  ) would choose to keep critical signal paths short, not limited to:- lower chance of injury/damage severing connection - smaller resource cost for producing the link (calories spent on neurons are calories not spent eating/procreating) - and of course latency (which is what your line of thinking boils down to, since there isn't a protocol spec, and coupling shouldn't be an issue) At the end of the day, I can't even conceive of a compelling reason not to collocate sensors with processing in an animal, so the costs and risks easily outweigh any hypothetical benefits.
|
|
#
?
Apr 9, 2020 04:35
|
|
|
maybe the metalworking thread can help figure out how to turn the magnesium into something thicker
|
|
#
?
Apr 9, 2020 06:30
|
|
|
asdf32 posted:LDO has become a generic term for regulator even though most regulators arn’t used in actual low dropout applications. To me LDO means "a linear voltage regulator that's newer than the 78xx range"
|
|
#
?
Apr 9, 2020 06:54
|
|
|
ante posted:That's part of it - Shorter distance to travel, faster the signal gets there. That would be the case if you put a DDR4 chip on there, for example. KnifeWrench posted:There are a number of reasons an intelligent designer (
|
|
#
?
Apr 9, 2020 12:59
|
|
|
Dominoes posted:That was an open question I had re the care on the PCB, then a long, looping cable after. Short answer is, this is why high-speed cables are expensive. Long answer is, inside the cable all the same rules are followed as your board. The individual wires are all the same length, they're twisted into pairs and the pairs are all surrounded on all sides by grounded shielding
|
|
#
?
Apr 9, 2020 13:54
|
|
|
taqueso posted:maybe the metalworking thread can help figure out how to turn the magnesium into something thicker I wonder if you could electroplate more magnesium on there? Never done it before, so it could be dumb.
|
|
#
?
Apr 9, 2020 16:55
|
|
|
taqueso posted:maybe the metalworking thread can help figure out how to turn the magnesium into something thicker i’ve lived in that thread for long enough at this point to brainstem-reflexively shout “MILL THE MAGNESIUM RIBBON INTO POWDER, MIX W FINE-MESH ALUMINIUM POWDER, CAST CUSTOM MAGNESIUM ELECTRODE ELEMENTS USING THE THERMITE PROCESS” in my sleep Stabby McDamage posted:I wonder if you could electroplate more magnesium on there? Never done it before, so it could be dumb. more or less, yes, but it’s a v obscure process that involves bizarre electrolytes/reagents. so far as i know NASA basically pioneered magnesium electroforming (like plating but for building up v thick structurally—useful deposits on a conductive electrode-mandrel) for the purpose of manufacturing extremely lightweight Mg-Al alloy parts with geometries impossible to achieve using normal sheet metal fab techniques. i’ve actually got the link to the research paper they put out on it in the 70s https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19710009381.pdf Ambrose Burnside fucked around with this message at 17:29 on Apr 9, 2020 |
|
#
?
Apr 9, 2020 17:21
|
|
|
Ambrose Burnside posted:i’ve lived in that thread for long enough at this point to brainstem-reflexively "Mill magnesium into a powder yourself" sounds like a fun way to blow up your mill and melt a hole through your shop floor  Admittedly most of my experience with powdered metals comes from a period in my teens where I was making my own fireworks thanks to my awesome and irresponsible dad being all "it's fiiiine, he knows how to not kill himself probably". In that context, magnesium and aluminum powder are things you use to make the whole process More Exciting, and I quickly learned a nice healthy respect for powdered metals in general and especially their occasional tendency to self-ignite.
|
|
#
?
Apr 9, 2020 17:31
|
|
|
oh yeah i dont even store 50 grams of magnesium ribbon w/o some forethought towards miraculous autoignition, multiplying its surface area by 1000x ain’t in my day planner, i just think it’s vaguely ridiculous that actually yes i do have an answer to the “best” way to produce thick magnesium parts exclusively from whatever’s in my garage due to pandemic lockdown, like an exceptionally suicidal version of junkyard wars or sth (also i’ll try to limit posting the barely-electronic-relevant stuff like the electrode spring bending itt, it’s not what y’all are here for)
|
|
#
?
Apr 9, 2020 17:49
|
|
|
Literally half my job is blowing up electronics in new and exciting ways
|
|
#
?
Apr 9, 2020 18:05
|
|
|
Ambrose Burnside posted:(also i’ll try to limit posting the barely-electronic-relevant stuff like the electrode spring bending itt, it’s not what y’all are here for) I mean I can't speak for everyone but it's exactly the kind of weird electronics-adjacent poo poo that keeps this thread fresh
|
|
#
?
Apr 9, 2020 18:21
|
|
|
Shame Boy posted:I mean I can't speak for everyone but it's exactly the kind of weird electronics-adjacent poo poo that keeps this thread fresh So much hobby poo poo now is "watch my Arduino blink a light" or "I 3D printed an anime model someone else made" I love this content
|
|
#
?
Apr 9, 2020 18:24
|
|
|
shovelbum posted:So much hobby poo poo now is "watch my Arduino blink a light" or "I 3D printed an anime model someone else made" I love this content That's always what hobby poo poo has been. "Check out this radioshack kit I bought and followed the instructions to make"
|
|
#
?
Apr 9, 2020 18:40
|
|
|
ante posted:Literally half my job is blowing up electronics in new and exciting ways Get some magnesium in there 
|
|
#
?
Apr 9, 2020 18:51
|
|
|
Shame Boy posted:"Mill magnesium into a powder yourself" sounds like a fun way to blow up your mill and melt a hole through your shop floor just build an airtight glove box around your mill and fill it with argon. yeesh. do i gotta think of everything around here
|
|
#
?
Apr 9, 2020 21:14
|
|
|
Cojawfee posted:That's always what hobby poo poo has been. "Check out this radioshack kit I bought and followed the instructions to make" I mean it gets kind of annoying to see after a while, but we all had to start somewhere. I agree that some of those kits can be a little too paint by numbers.
|
|
#
?
Apr 9, 2020 21:58
|
|
|
Cojawfee posted:That's always what hobby poo poo has been. "Check out this radioshack kit I bought and followed the instructions to make" the lovely end of it now tho is perfect content creator stuff, look what i made with no day job in my infinibudget sponsored shop with never-depicted assistants, the weird grognard poo poo like this is whats lacking
|
|
#
?
Apr 9, 2020 22:02
|
|
|
Infinite budget videos are pretty lame.
|
|
#
?
Apr 9, 2020 22:20
|
|
|
shovelbum posted:the lovely end of it now tho is perfect content creator stuff, look what i made with no day job in my infinibudget sponsored shop with never-depicted assistants, the weird grognard poo poo like this is whats lacking My favourite kind is the YouTuber who's been doing it for years. At this point, they've built up a huge body of knowledge but all of it without best practice, so they just make things hilariously hard for themselves for no reason. The best example I can think of is what's his name, James Bruton. The guy makes undeniably sophisticated robots, but he only ever uses Arduinos. All his programs are just mega loops, full of repeated code because what are functions? If he needs threading, he just physically adds more Arduinos. It is insane, hilarious, but also a little bit sad. I too am enjoying the weird battery stuff, that's the sort of obscure knowledge I love these threads for! Please keep posting
|
|
#
?
Apr 9, 2020 23:03
|
|
|
Yeah, I'm actually starting to flip around on my opinion of James Bruton. I started out kinda looking down on him. His stuff is really hilariously simple, just Arduinos and hobby servos and sometimes expensive hobby motor drivers, and poo poo tons of 3D printed brackets. But after a long time it starts to become complicated and impressive after going so deep down the rabbithole of adding more and more poo poo. He just keeps plowing forward. That might be the only way to accomplish that kind of thing on a weekly youtube schedule, though. My professional stuff has always been weeks and weeks of just pushing pixels on a screen, before hitting The Button and something impressive coming out of the manufacturing department. Doesn't make good entertainment.
|
|
#
?
Apr 9, 2020 23:08
|
|
|
I kind of feel sorry for him, his made a million iterations of that dog robot and he's so close to something cool but his basic design techniques are so limited he always just misses the mark and has to start from scratch again. His motivation is incredible, I would have given up countless times in his shoes. Our professional approach may not be entertaining, but at least our prototypes work (or at least work well enough that they can be fixed up without a total scrapping of the design). Poor James could just do the maths and work out if his motors are powerful enough before building anything, but he never does! It's exhausting!
|
|
#
?
Apr 9, 2020 23:18
|
|
|
taqueso posted:You can still patent bent wire if you do it right. You just need to combine it with something else. For example, I was told by a SRAM EE that Shimano has a patent on parallel stacks of PCBs on a bicycle so sram used perpendicular PCBs in their e-shifter design.
|
|
#
?
Apr 9, 2020 23:25
|
|
|
|
| # ? Apr 16, 2024 20:48 |
|
|
It is absolutely an abuse of the patent system and a perversion of the entire concept
|
|
#
?
Apr 10, 2020 00:07
|
|