|
I want to make a sign with letters spelled out in LEDs, and be able to blink individual letters with an AVR. That means driving banks of bright LEDs with total currents higher than what an AVR pin should do. All my LED stuff so far has been "apply voltage, use appropriate resistor, problem solved", but I think I need to learn about LED drivers to do this, right? I'd prefer not to end up with a discrete transistor and resistor for every bank of LEDs... I don't know anything about this, though, including terminology. Here's what I'm thinking:  What is the mysterious object in red, and where can I get a cheap one from China? (I believe that all my projects should be made out of parts cheap enough so I don't want to scavenge them after it's done.)
|
#
?
Nov 6, 2013 23:21
|
|
|
|
| # ? Jun 3, 2024 22:14 |
|
|
I recently used an eval board for something like that: http://www.ti.com/tool/tps61150evm-150 The eval board isn't cheap, but I'm sure you could get the core component cheap enough you wouldn't want to salvage it. TI makes a lot of LED driver solutions, the other one in my browser history for whatever reason is this but I don't remember working with it.
|
|
#
?
Nov 6, 2013 23:38
|
|
|
TasogareNoKagi posted:Wish I thought to check here before I went and bought all this stuff The crimp spec isn't lying about 22 AWG, I've used similar crimps from Molex and that's how it's supposed to look. Also you'll need a special (read: expensive) hand crimper that works specifically on that family of crimps. The crimp tool you bought works only with lug-type terminals such as spade and ring terminals.
|
|
#
?
Nov 7, 2013 00:53
|
|
|
If you don't expect to use the terminals again, grab an appropriate needle-nose pliers and crimp them manually. Crimp tools can be a bit too expensive for one-offs.
|
|
#
?
Nov 7, 2013 01:20
|
|
|
Stabby McDamage posted:I want to make a sign with letters spelled out in LEDs, and be able to blink individual letters with an AVR. That means driving banks of bright LEDs with total currents higher than what an AVR pin should do. All my LED stuff so far has been "apply voltage, use appropriate resistor, problem solved", but I think I need to learn about LED drivers to do this, right? I'd prefer not to end up with a discrete transistor and resistor for every bank of LEDs... http://www.digikey.com/product-search/en/integrated-circuits-ics/pmic-led-drivers/2556628 Filter by DIP and constant current. Some of those even have internal shift registers, which is pretty sweet.
|
|
#
?
Nov 7, 2013 01:40
|
|
|
EE/CEs: have any good interview questions for a fresh graduate? (GA Tech) I've got some I've thought up about basic FPGA stuff on the CE side of things (interconnects, pipelining, fan-out, etc), but I think my EE questions are a bit skewed for someone fresh out. What are some good ones to check basic concepts?
|
|
#
?
Nov 7, 2013 04:36
|
|
|
TasogareNoKagi posted:Wish I thought to check here before I went and bought all this stuff Like Boy With Stick said, these are the wrong crimpers for the job. I've crimped these exact contacts with this crimper: http://www.adafruit.com/products/1213 Not exactly correct, but they meet the wire pull-out spec using the tester at work.
|
|
#
?
Nov 7, 2013 14:29
|
|
|
I sent a resume to a company recently and they replied saying to send them some "representative examples of prior work". I'm not really sure how to respond to this - certainly I can't send them design work I've done under hire at previous jobs, and my hobby work doesn't amount to much more than scribbles in a notebook. Given that I fall more on the theoretical side of things there isn't really very much I can show them.
|
|
#
?
Nov 7, 2013 17:13
|
|
|
Are there any products that are currently out on the market that you worked on? Most NDAs are okay with you saying that you worked on something after it's put on sale, as long as you aren't revealing trade secrets or whatever. Then again if you were working for a company as an outside contractor they may not want that fact publicized so I dunno. It's always hard to balance. Part of why I try to maintain an interesting side practice of my own projects, so that I always have something to talk about and show if the need arises.
|
|
#
?
Nov 7, 2013 22:49
|
|
|
Cyril Sneer posted:I sent a resume to a company recently and they replied saying to send them some "representative examples of prior work". I'm not really sure how to respond to this - certainly I can't send them design work I've done under hire at previous jobs, and my hobby work doesn't amount to much more than scribbles in a notebook. Given that I fall more on the theoretical side of things there isn't really very much I can show them. Companies are interested in two things. If you have professional experience in industry, they want to know about products you've developed or services you've provided (you don't have to mail them the hardware and code or whatever, just a description). If you're coming from academia, then they want to see publications. If you can't do either of those, then you are probably out of luck unless you are well-connected or otherwise can get a personal interview. That's why publishing is crucial for grad students. It's proof that you did some poo poo and understand it. Scribbles are never going to impress, no matter how awesome the scribbles are. As an anecdote, a PhD grad who went through my program several years back got hired by a company, partially because he told them he had published a manuscript in a certain journal. Shortly after starting the job, it came to light that he had submitted it, but it hadn't actually been accepted yet. His boss flipped and told him to immediately finish the manuscript and get it published or he would be fired (he did finish it eventually). It might seem absurd, but there is good logic to it. The boss wasn't suddenly doubting the guy's intelligence, but boss still had to know that he was capable of actually completing a large task, right up to the end. That's hugely important in industry, and no amount of scribbles will ever compensate for one solid, thoroughly developed body of work. ANIME AKBAR fucked around with this message at 00:00 on Nov 8, 2013 |
|
#
?
Nov 7, 2013 23:52
|
|
|
The rule of thumb I've always heard is you can say what you did, you can't say how you did it. Most of my EE interviews are discussing the products they shipped and figuring out exactly what part of it the candidate was responsible for.
|
|
#
?
Nov 8, 2013 00:01
|
|
|
I understand why they would ask for something like that; that's not the issue. It's how do *I* reply given that I'm not and never have been a design engineer (I'm more of an Applications Specialist / Testing & QA guy). Send them a requirements and specifications document I wrote? A new testing procedure my company implemented? edit: I mean, it could just be that they want a design engineer and I'm not the right match, which is fine, but it's not like I'm going to shoot off an email saying to withdraw my application 
Cyril Sneer fucked around with this message at 04:21 on Nov 8, 2013 |
|
#
?
Nov 8, 2013 04:17
|
|
|
"We built this machine that punches people in the face through the internet. My role was to come in at the end and evaluate how effective we were, and assist with prototype modifications. I tested it for velocity at impact point, energy consumption, and appropriateness of the situation. Here is a public picture from our web store. Contact me if you have any questions."
|
|
#
?
Nov 8, 2013 04:36
|
|
|
So I bought a little frequency generator kit. It's this one from Sparkfun or the same thing from Amazon for $4 less and prime-eligible. Thought I'd give a little trip report.  So far it seems fine. The kit was super easy to put together, even for a soldering greenhorn like me. It took me a couple hours to put together, but that's spending the first half of the assembly getting my solder technique down and the entirety of the assembly in teleconferences. Certainly it was some good soldering practice. By the end, I was zipping over the pins for the buttons at a pretty good click with consistent and quick solder work, it definitely upped my comfort level. Effectively all the soldering involved was through-hole, as all the smd components were already taken care of. Some of the sparkfun comments complained about soldering on the bnc connector, but by the time I got to that one, it was fine, so I don't know what people were complaining about. Once assembled, it looks nice and seems solid enough. The controls are clear and easy to use. The screen is nice and bright and easy to read. The unit is nice and compact, and comes with everything required for basic use (all board components, power supply, and a bnc connector with alligator clips.) It's actually a fair amount more featured than sparkfun claims. There was a firmware update that added a number of features, and from the comments (and my own unit) it appears all units are shipping with the updated firmware now. Specifically, the frequency range is up into the megahertz rather than the 200khz stated by sparkfun. Also, in addition to the noted sine, triangle, and square wave output, it will also do +/- ramps, +/- stairs, and can use user generated functions, I believe loaded via the miniusb port. I haven't gotten to examine the output yet aside from confirming that it's functional. Next week I'll take it down to the makerspace and plug it into the scope there: it's a fairly high quality scope and should give me some good info. Supposedly the output gets a bit fuzzy as you go into the mhz range or take the amplitude down to the mV range. We'll see what the scope has to say about it. As a basic generator for a hobbyist who just needed something quick, so far it seems more than sufficient. At the very least it was a fun little bit of assembly and it'll come in handy for a couple projects I have going in the immediate future. And the price was just fine, too. Yes, a person could make a comparable one for cheaper, but if you're like me, there's no telling when it'd actually come together, if ever. So for m, this was just the ticket. 
|
|
#
?
Nov 8, 2013 04:43
|
|
|
Sagebrush posted:Are there any products that are currently out on the market that you worked on? Most NDAs are okay with you saying that you worked on something after it's put on sale, as long as you aren't revealing trade secrets or whatever. This probably varies a lot depending on what market you're in, I'm under a strict NDA (and it's not a joke one like when I worked at a department store) about all internal company matters, including what I have/will work on and our products in general.
|
|
#
?
Nov 8, 2013 19:27
|
|
|
Serious question: What qualifications are necessary to be one of the guys who creates the weird dildo machines in alternative porn? I mean that's at least entry level robotics and circuits, right?
|
|
#
?
Nov 8, 2013 20:46
|
|
|
It's not hard to mount a dildo on a power drill, dude
|
|
#
?
Nov 8, 2013 21:09
|
|
|
What porn are you watching that the machine isn't something that is picked up at a hardware store and then a dildo just attached to it?
|
|
#
?
Nov 8, 2013 21:14
|
|
|
The better ones, apparently.  My CMOY components came in from Mouser today. Gonna make my girlfriend a headphone amp today! 
|
|
#
?
Nov 8, 2013 22:37
|
|
|
Xovaan posted:Serious question: What qualifications are necessary to be one of the guys who creates the weird dildo machines in alternative porn? I mean that's at least entry level robotics and circuits, right? These guys are fabricators and/or machinists. Nothing but mechanical linkages and electric motors. The electronic sex toy equivalent would be TENS/e-stim machines and such. Please don't attempt to make these.
|
|
#
?
Nov 9, 2013 00:31
|
|
|
sixide posted:These guys are fabricators and/or machinists. Nothing but mechanical linkages and electric motors. Or biofeedback devices  Who wouldn't want to couple vibration strength to the output of a perineometer? Delta-Wye fucked around with this message at 00:55 on Nov 9, 2013 |
|
#
?
Nov 9, 2013 00:53
|
|
|
Delta-Wye posted:Or biofeedback devices Someone who doesn't want their genitals to be over stimulated to the point orgasm is hard to achieve for a while afterwards.
|
|
#
?
Nov 9, 2013 01:19
|
|
|
So basically not the Sybian market share!
|
|
#
?
Nov 9, 2013 01:26
|
|
|
Terminal Entropy posted:Someone who doesn't want their genitals to be over stimulated to the point orgasm is hard to achieve for a while afterwards. A little stronger than I had in mind, but I like the way you think 
|
|
#
?
Nov 9, 2013 01:29
|
|
|
Gets mention on Love Line when someone goes to town and ends up de-sensitized. Though a couple of weeks ago a guy called in during open forum about a DIY electronic muscle exorciser to have a fun night. EDIT: Don't think I have seen a better usage for that smilie until now. Any suggestions to improve the structure of this:  I might go back and try to heat shrink and epoxy everything down as my initial failed plan was a combination of heat shrink and wire hanger. Terminal Entropy fucked around with this message at 01:37 on Nov 9, 2013 |
|
#
?
Nov 9, 2013 01:30
|
|
|
Bad Munki posted:So I bought a little frequency generator kit. It's this one from Sparkfun or the same thing from Amazon for $4 less and prime-eligible. Thought I'd give a little trip report. Thanks for this, I've been looking for a cheap-o function generator in the <1MHz range so I just ordered it. Since you have one sitting in front of you, how hard do you think this would be to put in a box? It looks like the front plate is held in place with some spacers, would it be possible to disconnect those and run them through the front of a case as a bulkhead-style mount, or is the front panel connected to the main board through headers? e: Did the wall wart power supply come with it? e2: To phrase my mounting question in a slightly more clear way - could the front panel be moved out about 1/8" to account for the thickness of a box wall, or is it physically tightly coupled to the main board? PDP-1 fucked around with this message at 02:46 on Nov 9, 2013 |
|
#
?
Nov 9, 2013 02:36
|
|
|
Comes with the board and all components, a bnc to alligator patch cable, and the required 15V power supply. It'd be extremely easy, and I plan to do so with mine. However: the bnc connector goes through the front plate and is then soldered to the board in the default configuration. The standoffs provide the right separation (guessing exactly 1/2", but I can't check right now since I'm about a hundred miles from my workbench.) An easier way would be to get slightly longer screws for mounting the faceplate to the standoffs and have your enclosure go in front of the faceplate with a suitable rectangular cutout for the whole faceplate to be exposed. The way the bnc connector goes on, it has a lip in front of the faceplate, so there's not much play. When I get home, I'll get some pics for you so know what to expect. Like I said, it'd be super easy to enclose, but without replacing the entire faceplate, you'll probably need to mount the unit in the enclosure a certain (perfectly acceptable imho) way. And to clearly answer your edit question: no, there's definitely not 1/8" of play. But you could get around that by putting the enclosure in front of the faceplate instead of behind, or perhaps behind and replacing the front standoffs with 3/8" standoffs but you'd have to solder the BNC connector with your additional front panel in place (not really a problem.) Don't forget the travel needed by all those buttons. They protrude about 1/8" and need that full 1/8" travel to faceplate-flush for use. Bad Munki fucked around with this message at 03:14 on Nov 9, 2013 |
|
#
?
Nov 9, 2013 03:07
|
|
|
Thanks! Mounting the front plate behind the enclosure wall would be plenty good enough - I'm looking for functional here, not pretty. I'm just not a big fan of random small boards flopping around my bench. I'll likely also move the power input to a through-wall connector, or maybe just thread the wire through a hole and do something to keep it from getting tugged out of the board socket. It's amazing how much prices for test equipment are falling now that pretty much anyone can design a PCB and get a few hundred populated copies knocked out. Sure, it's not the most amazing gear in the world but it is plenty good for 99% of hobby work.
|
|
#
?
Nov 9, 2013 03:35
|
|
|
Terminal Entropy posted:Any suggestions to improve the structure of this: Uh...what is it, exactly? A right-angle USB adapter?
|
|
#
?
Nov 9, 2013 04:57
|
|
|
Yup. I have my mouse connected through my keyboard, but space is limited on the keyboard tray.
|
|
#
?
Nov 9, 2013 08:27
|
|
|
Terminal Entropy posted:Yup. I have my mouse connected through my keyboard, but space is limited on the keyboard tray. One of these would certainly improve the structure: http://www.l-com.com/usb-flexusb-adapter-type-a-male-female
|
|
#
?
Nov 9, 2013 13:50
|
|
|
Terminal Entropy posted:Gets mention on Love Line when someone goes to town and ends up de-sensitized. This is a job for InstaMorph/ThermoMorph/Friendly Plastic/Fantastic Plastic/whatever brand of Polycaprolactone is cheapest. http://www.amazon.com/Thermomorph-Moldable-Plastic-17-8-Molding/dp/B00D3LAZ9O/ref=sr_1_5?ie=UTF8&qid=1384018266&sr=8-5&keywords=instamorph You warm it to 60 C in water on a stove top, pull it out, and mold it with your hands into whatever you want. I used it to finish a little LED light I helped some kids make:    And I used it to turn a 9V battery clasp into a mountable battery holder:     I've also made microphone stands, LED sculptures, and other stuff with it, and it's only like $16 for a pound of it (all of this stuff has used up less than a third of my first jar).
|
|
#
?
Nov 9, 2013 18:38
|
|
|
https://www.youtube.com/watch?v=SMgDfkqn4n0 It is great for slapping together a lovely prototype really quick, especially with an awkward shape.
|
|
#
?
Nov 9, 2013 19:10
|
|
|
Stabby McDamage posted:This is a job for InstaMorph/ThermoMorph/Friendly Plastic/Fantastic Plastic/whatever brand of Polycaprolactone is cheapest. Will have to look around at that, is looks perfect.
|
|
#
?
Nov 9, 2013 20:00
|
|
|
Holy poo poo, this wireless power thing actually worked on the very first try. Didn't even have to fiddle with the frequency, it just worked. Here's a little disk with an LED on it:  and if I move it on to the X...  Tah-dah! I have the exact same device underneath, but with a couple wire leads instead of an LED. They're hooked up to my frequency generator for now. I just used the same board because they come in threes from OSH Park, so it was a convenient transmitter coil that I knew would have the same characteristics.  Here's the whole shebang for the moment:  When I ordered the boards, I accidentally made the vias in the middle too small for the LED leads, and there wasn't enough play to drill it out, so I just soldered the LED on top, which is why it's standing so far off. Also, using the same board as the power source meant the active region was fairly small. Once I've worked out some kinks, I'll see about making a larger pad to work with, not sure what sort of hiccups I'll run into at that point so I'm going to take it one step at a time. Still, though: prototype works. https://www.youtube.com/watch?v=cbIWDnGJ9_s e: Oh, also: THANKS ANIME AKBAR, I'm pretty sure that without your awesome guess at the inductance of this thing, I would have been screwing around for ages trying to get it to work. Thanks to your insight, I just had to solve for a couple other numbers and call it a day. 
Bad Munki fucked around with this message at 03:15 on Nov 10, 2013 |
|
#
?
Nov 10, 2013 02:53
|
|
|
Ughf 
|
|
#
?
Nov 10, 2013 03:09
|
|
|
Follow it up by doing this next: http://chrisrieger.com/projects/levlight
|
|
#
?
Nov 10, 2013 04:14
|
|
|
Cool project! I'd love to wire those into the underside of a desk or something for wireless light on the surface. I'm learning about switches right now for this CMOY and I'm attempting to install this: http://www.alps.com/WebObjects/catalog.woa/E/HTML/Potentiometer/RotaryPotentiometers/RK097/RK0971221Z05.html Schematic: http://tangentsoft.net/audio/cmoy-tutorial/misc/cmoy-tangent-sch.pdf You can see it here, but with the other switch installed instead (green box is pot):  And here's the way it looks assembled: http://tangentsoft.net/audio/cmoy-tutorial/misc/mini-cmoy-layout-0150.pdf My question: How do I know which is the right way to install the red (+) and black (-) wires to the potentiometer?
|
|
#
?
Nov 10, 2013 18:20
|
|
|
It looks like that part has a type 3B taper, which means it's a linear potentiometer, right? If you don't know, you can check to be sure by measuring the resistance. If it's linear it will be exactly half the rated value when the dial is centered. If it is a linear pot, then electrically the arrangement of the outer wires doesn't matter -- as long as you have "IN" on one side, ground on the other and C2 connected to the middle pin, it'll work. Reversing the outer wires will reverse the direction of the dial. Sagebrush fucked around with this message at 19:00 on Nov 10, 2013 |
|
#
?
Nov 10, 2013 18:56
|
|
|
|
| # ? Jun 3, 2024 22:14 |
|
|
It clicks off when you turn it completely CCW-- there's eight pins coming off the potentiometer. I followed his wiring for my own wiring, but I'm using that potentiometer as my on/off switch on pins 7 and 8 instead of his toggle switch he has in that image, which are almost covered by the red and black wires, respectively. Does it matter which way they're hooked up?
|
|
#
?
Nov 10, 2013 19:17
|
|