|
ANIME AKBAR posted:Yeah it might be fun until you have to do it in IEEE 754 floating point. If it's floating point, definitely spring for something with an FPU: http://focus.ti.com/docs/prod/folders/print/tms320f28335.html What kind of resolution are you looking at? I'm currently doing some CT reconstruction stuff in Matlab (filtered back projection). It doesn't exactly scale well, my reconstruction operation has to be performed resolution*resolution*slice count times. Runs about 30 seconds for 1024x1024 * 30 slices on my i7.
|
#
?
May 30, 2010 01:48
|
|
|
|
| # ? Jun 10, 2024 23:42 |
|
|
Retro-Future Rodent posted:Cool, once the weather clears I'll have to take a trip down to my own. i've actually already built a much more complicated solartracker, which i'll release the shematics for once my exams are over  mebbe i should look into some BEAM stuff, i have a whole book that covers that subject, and it's an awesome book.
|
|
#
?
May 30, 2010 10:29
|
|
|
catbread.jpg posted:If it's floating point, definitely spring for something with an FPU: quote:What kind of resolution are you looking at? I'm currently doing some CT reconstruction stuff in Matlab (filtered back projection). It doesn't exactly scale well, my reconstruction operation has to be performed resolution*resolution*slice count times. Runs about 30 seconds for 1024x1024 * 30 slices on my i7.
|
|
#
?
May 30, 2010 13:19
|
|
|
I have an unidentified metal that solder absolutely refuses to stick to, and I need to solder some wires on to it. I can try to identify the metal, if that would help, but is there anything I can try first to make the solder stick?
|
|
#
?
Jun 1, 2010 05:09
|
|
|
Dooey posted:I have an unidentified metal that solder absolutely refuses to stick to, and I need to solder some wires on to it. I can try to identify the metal, if that would help, but is there anything I can try first to make the solder stick? Flux
|
|
#
?
Jun 1, 2010 05:41
|
|
|
ANIME AKBAR posted:A while back, I was actually looking at this exact family of MCUs (though it was the piccolo, not the delphino) as a step up from atmel's stuff, but I never found a decent application so it's just been gathering dust. It would probably work well for this application, but I doubt I could master it. Where do you think the difficulty would lie? They're not that bad.
|
|
#
?
Jun 1, 2010 10:26
|
|
|
Scarboy posted:Flux Didn't help. And I used lots of it.
|
|
#
?
Jun 1, 2010 17:08
|
|
|
Dooey posted:Didn't help. And I used lots of it. Several unrelated ideas: If it's aluminum, you might have some trouble due to the oxide layer. Are you sure there's no clear enamel coating on it? Rub it with sandpaper to make sure you get off any surface corrosion/coatings, then try flux+solder. How big is the piece of metal? How much heat are you using? Soldering to a chunk of conductive metal is tough because the heat tends to conduct away from your solder joint.
|
|
#
?
Jun 1, 2010 17:35
|
|
|
SnoPuppy posted:How big is the piece of metal? How much heat are you using? Soldering to a chunk of conductive metal is tough because the heat tends to conduct away from your solder joint. This is almost certainly the problem, as I did fine soldering to a much smaller piece of the same material. So I guess I just need to heat it up longer?
|
|
#
?
Jun 1, 2010 19:05
|
|
|
Dooey posted:This is almost certainly the problem, as I did fine soldering to a much smaller piece of the same material. That, or use a lot of heat in a short time (blow torch!)
|
|
#
?
Jun 1, 2010 19:54
|
|
|
catbread.jpg posted:Where do you think the difficulty would lie? They're not that bad. I never properly learned C. Assembly was fine for simple MCUs like avrs, but obviously wouldn't cut it for C2000 core parts. When I was messing around with my devkit, just looking at the example header files would give me a headache (about ten files, a few hundred lines of code). Hopefully at some point I'll have enough time to take a proper programming course that uses C (our required undergrad course taught java of all things...), so I can properly understand things.
|
|
#
?
Jun 2, 2010 13:09
|
|
|
ANIME AKBAR posted:I never properly learned C. Assembly was fine for simple MCUs like avrs, but obviously wouldn't cut it for C2000 core parts. When I was messing around with my devkit, just looking at the example header files would give me a headache (about ten files, a few hundred lines of code). Hopefully at some point I'll have enough time to take a proper programming course that uses C (our required undergrad course taught java of all things...), so I can properly understand things. Although my anecdotal sample size is pretty small, I have never encountered an EE whose school offered an appropriate C course for Embedded programming. My Intro to Programming class was C, but half the students in the course were Psych majors. It was almost worse than useless.
|
|
#
?
Jun 2, 2010 14:36
|
|
|
My school had fantastic courses for C, my favorite one being Microprocessor Systems and Interfacing that focused on analog/digital signals and had several labs involving embedded programming. Went through everything from MCU hardware architecture, peripheral interfacing, to practical stuff like grounding and shielding. That class alone got me super interested in embedded programming to the point where all my co-op jobs after it involved microprocessors. We had other classes where we'd use microprocessors in projects and stuff too. These were all third year or higher classes though; I'm not sure why you would expect an intro class to be anything but poo poo. For reference a good chunk of my job now is embedded programming, and I work with freescale DSPs and microchip MCUs (freescale sucks, microchip rules)
|
|
#
?
Jun 2, 2010 14:56
|
|
|
The only programming course open to EEs other than Intro to C is a Computer Interfacing course which teaches Assembly on an ancient, half-broken suite of MC68HC11 development equipment, unless you count some simple intro to Labview or Matlab courses. As a graduate student, its often taken for granted that the students are able to fluently program in *some* language or another, and it's hurting a lot of my less self-starting peers (although a lot of these people probably shouldn't be in graduate school, but that's another can of worms).
|
|
#
?
Jun 2, 2010 16:05
|
|
|
Zuph posted:Although my anecdotal sample size is pretty small, I have never encountered an EE whose school offered an appropriate C course for Embedded programming. My school taught us C, the closest we got to embedded programming was implementing simple FIR filters on DSP chips. Still laid the groundwork though. We also learned a few assembly dialects through processor design courses.
|
|
#
?
Jun 2, 2010 16:11
|
|
|
My intro C class had an average mark of 35%. I accidentally slept through a quiz(6% off right there), showed up for the (2 hour) final hald an hour late, left the final half an hour early, and still pulled off an 89%  I was one of the only people that already knew C, that class was awesome.
|
|
#
?
Jun 2, 2010 16:32
|
|
|
We were thinking of setting up a greenhouse soon and my mind wandered to greenhouse automation. I have some experience with microcontrollers. I don't do embedded software for a living but I do a ton of assembler--mostly x86, but I've dealt with z80 briefly, HC12, and 68k. For any microcontroller with a limited number of registers I'd much prefer to code in a C since I am not fond of hand coding out all the juggling. Some important things to worry about is watering, air temperature, plant temperature, humidity, and light. Beyond that I'd like some sensors to pick up on when things are wrong. For example, I'd like alarms if it thinks something failed. A water pressure sensor should be able to trigger something to notify me in some way if the watering system appears to be messed up. Ultimately I'd like to mess around with all facets so I need an appropriate amount of I/O ports to do the job. Chances are in all cases it will involve a device that's either on or off, so single I/O pins will probably do it. I can put together basic hardware to deal with that; it would probably just involve relays. Where I need some help is in selecting a controller. I want something that is remotely accessible and can withstand humidity and potentially getting wet. I am close to just using an Atom-based system running off an SD-MMC card and hijacking a serial port to toggle pins. It looks like that would be about $200. There I could give it some really simple commands from time to time and it could do all the thinking and monitoring. Alternately I was thinking about microcontrollers with ethernet or USB ports that weren't smart at all but could handle commands to turn stuff on or off with something else remotely calling the shots. I wondered generally how robust these systems are and how easily it is to implement communications with them. I had some bad experiences recently trying to get a USB stack working on something that I'm inclined against using one; I'd love something that had enough support to do socket I/O with perhaps just the inclusion of some existing source.
|
|
#
?
Jun 2, 2010 16:55
|
|
|
You can get a development board with a nice microcontroller from a number of big name companies for cheap (like this is 75 bucks and is more powerful than anything you'll need), which would come with a programmer, power supply, and an easy to use board. No need to go "dumb" when ridiculously fast micros cost <10 bucks a piece nowadays. Then if you want easy, reliable networking, I've used Xport's ethernet modules and they're incredibly braindead easy to set up. Basically connect to your micro's serial pins and power/ground (3.3V), and plug in an ethernet cable. It'll automatically get an IP from your network and you're good to go. The module is 50 bucks and will probably save you a lot of headache. It also has a small embedded web server so you can even just write up a webpage and view your greenhouse stats through a browser anywhere (disclaimer: I haven't used the embedded web server, I just used it as an ethernet -> serial bridge). edit: these things will withstand standard industrial temperature range (-40 to 85) and for humidity there are a number of coating sprays you can purchase Zo fucked around with this message at 17:07 on Jun 2, 2010 |
|
#
?
Jun 2, 2010 17:04
|
|
|
Zo posted:You can get a development board with a nice microcontroller from a number of big name companies for cheap (like this is 75 bucks and is more powerful than anything you'll need), which would come with a programmer, power supply, and an easy to use board. No need to go "dumb" when ridiculously fast micros cost <10 bucks a piece nowadays.
|
|
#
?
Jun 2, 2010 18:19
|
|
|
Rocko Bonaparte posted:At $125 I would have thought that fine a year ago, but having to work with Atom-based systems for about two years now and seeing what they can do, it looks like I can get one together with case and RAM for $200. And there I'd have most everything built in except for some pins to toggle relays. They don't need a fan to cool them--especially without the rigor I intend to apply. They'll go through a lot more power but they're much better than a regular PC chip. It could still be a stupid idea for reasons I wouldn't know though. Perhaps affixing the USB I/O would introduce weak points for moisture and heat. Or maybe I'd get stuck with reading sensors since it wouldn't have A/D converters. Yeah can't say I have much experience with ATOM-based stuff. Too close to PCs for me, and way too much stuff I'll never use (graphics, external RAM, etc). If you're experienced with them already go for it, but application-wise a microprocessor is probably better suited for this kind of stuff, though not worth it if you have to go through a steep learning curve. Also, having ADCs would definitely help with some of the sensors.
|
|
#
?
Jun 2, 2010 18:31
|
|
|
Rocko Bonaparte posted:At $125 I would have thought that fine a year ago, but having to work with Atom-based systems for about two years now and seeing what they can do, it looks like I can get one together with case and RAM for $200. And there I'd have most everything built in except for some pins to toggle relays. They don't need a fan to cool them--especially without the rigor I intend to apply. They'll go through a lot more power but they're much better than a regular PC chip. It could still be a stupid idea for reasons I wouldn't know though. Perhaps affixing the USB I/O would introduce weak points for moisture and heat. Or maybe I'd get stuck with reading sensors since it wouldn't have A/D converters. You'll probably have environmental issues (heat/humidity) with either option. It might be easier with a micro though, because the will almost certainly run cooler. Not to mention that getting the IO you need for a PC will probably be harder or more expensive than doing it with a micro. Actually, you could do a combination of the two - use a small micro to read sensors and control actuators and have it talk back to a host PC to do the data logging and higher level control.
|
|
#
?
Jun 2, 2010 19:17
|
|
|
SnoPuppy posted:You'll probably have environmental issues (heat/humidity) with either option. It might be easier with a micro though, because the will almost certainly run cooler. Micros will definitely not have an issue with either heat or humidity. They're (mostly) built for industry standards and will survive canadian winters and california summers, all outdoors. Spray it with a $2 can of PCB lacquer and you can dip it in a bucket of scalding saltwater.
|
|
#
?
Jun 2, 2010 20:43
|
|
|
ahahaha I took a look at what his PhD students have set up and they're using some insane 16 channel, 750Mbps digital synthesizer interfacing with matlab in order to do pulse shaping (not sure how they do sampling). Granted, my hardware is simpler and I only have much fewer coils, but I'm still working with the same frequency. I have no idea why this guy would think an arduino is applicable here.
|
|
#
?
Jun 3, 2010 12:59
|
|
|
Unless... you made an Arduino CLUSTER!!! (Okay yeah, I'm thinking FPGA)
|
|
#
?
Jun 3, 2010 18:12
|
|
|
ANIME AKBAR posted:ahahaha I took a look at what his PhD students have set up and they're using some insane 16 channel, 750Mbps digital synthesizer interfacing with matlab in order to do pulse shaping (not sure how they do sampling). Granted, my hardware is simpler and I only have much fewer coils, but I'm still working with the same frequency. I have no idea why this guy would think an arduino is applicable here. Hah! Yeah, that's what I was thinking... Some of the new FPGAs from Xilinx are *very* capable - I'd look into the Spartan 6 class. Generally you want the fastest DAC you can get, since it makes the analog filter design much less stringent (i.e. it's much, much nicer to have your max frequency be like 0.1 Fs than 0.4 Fs). That could be why they have that synthesizer - any idea who makes it? You should be able to get away with a slower ADC, especially if you mix down to baseband or low IF first. Mixing might not even be needed if your ADC has a wide front end and you can alias. I know Analog Devices and Linear Tech have a lot of good parts across the board.
|
|
#
?
Jun 3, 2010 20:38
|
|
|
For the output I know I'll be using a DDS to generate the carrier, and the output modulation will probably just be a square pulse (so more like a switch than a mixer). And the signal will be downmixed before sampling, so it will won't have a very high bandwidth. I'm expecting something in the range of 50-100KHz, which is still a lot for an avr.
|
|
#
?
Jun 3, 2010 23:21
|
|
|
Anime Akbar, were you the one selling those high power caps? If so, do you still have them? Hillridge fucked around with this message at 16:31 on Jun 4, 2010 |
|
#
?
Jun 4, 2010 02:59
|
|
|
I brought a AM1P-0524SZ. This will convert 5V to 24V DC. If I connect a 9V battery I get ~40V, will this damage it or is it ok to do this?
|
|
#
?
Jun 4, 2010 12:12
|
|
|
http://www.aimtec.com/site/Aimtec/files/Datasheet/HighResolution/AM1P-Z.PDF That part is designed to work on a supply between 4.5 and 5.5V. A little over won't hurt, but 9V is probably too much.
|
|
#
?
Jun 4, 2010 16:38
|
|
|
Hillridge posted:http://www.aimtec.com/site/Aimtec/files/Datasheet/HighResolution/AM1P-Z.PDF That would explain why it doesn't work any more  .
|
|
#
?
Jun 4, 2010 17:13
|
|
|
Speaking of things that no longer work, let us take a look at a power supply i've had hanging around on a shelf. This little gem was built by my mother sometime in the 90es as part of an introductory course to soldering and electronics. It worked perfectly until about two months ago, where i got curious to as how many amps it could deliver. My teacher told i could estimate this by hooking it up to a decade resistor (big long thing with tons of knobs to switch resistance with) and it promptly stopped working! It no longer delivers anything to the outputs and the voltmeter doesnt move either. what are the possible faults?    apparently the design they used in that course was somewhat modular, what with the control board for the whole thing being easily swappable. i guess this was because it was an intro course and people tend to be somewhat bad at soldering, so stuff could easily be hosed up. many of the solder joints inside this thing are horrible.
|
|
#
?
Jun 5, 2010 11:05
|
|
|
There are no fuses? The LM301AN can supply continuous short circuit current, so it's probably fine. None of your caps or resistors look fried or explodey. My bet's on that transistor. Does it have a part number? What are those other two black things?
|
|
#
?
Jun 5, 2010 11:33
|
|
|
There is a fuse, i have checked it, nothing's blown there. Big black thing next to huge cap: KBP-04 BY164 and then there's a marking + AC - Transistor next to weirdo resistor: BC147B L7827 Silver transistor: TFKBC140/10 there's also the power transistor chilling in a giant heatsink outside the case: 7845 2N3055H 6 these are the markings on the transistors and the weird black thing i have a bunch of lying around but havent bothered to google yet.
|
|
#
?
Jun 5, 2010 11:53
|
|
|
Well it's hard just looking at it. Plug it in and start probing the signal from the 120VAC and see where it stops.
|
|
#
?
Jun 5, 2010 14:13
|
|
|
I found a clock radio with its power cord yanked in a dumpster that I fancied fixing and using. Turns out the ripped out cord also ripped out a chunk of the PCB. I cant tell if that chunk was important or not but with it out, there's no labeling for where the power cord used to go. The 4 solder points visible before the break are for a transformer so I'd assume I connect power and ground to two points...but not sure which. Any help? Side question, any one ever solder together their own solar panel out of cells? Thinking about doing it as a budget way to get some solar power in my home; just wondering if there are any tips
|
|
#
?
Jun 7, 2010 04:37
|
|
|
So, I recently received quite possibly the best gift ever -- the entire contents of an electronics store that went bankrupt some time in the 1980s. It contains an ungodly number of resistors of any value you can imagine, thousands of inductors and capacitors, some big ol' electrolytics that I am too afraid of to use or even test, a number of old transistors and op amps, and an absolutely staggering quantity of integrated circuits (including some fun things such as original 8080s and related computer parts). Most of this I've had no trouble at all identifying. However, I'm still quite new to the world of electronics, so I've been unable to identify a good portion of the integrated circuits. I'm hoping someone might be able to point me in the right direction by looking at some of what I have. I'm pretty sure most of these are original 7400 series logic ICs, but I don't know how to go about relating the numbers to functions: (click for bigger)   These, however, have me at a complete loss:  Any pointers would be greatly appreciated.
|
|
#
?
Jun 7, 2010 06:34
|
|
|
Hey man. You wanna be my friend? Like, best buddies? Part numbers can be looked up on something like http://www.datasheetcatalog.com/ The number on the box will give you a good idea of what's inside, but the full part number on the chip itself will help you figure out if it's a hex inverter or SRAM, for example. edit: I'm serious, I'm so loving jealous right now.
|
|
#
?
Jun 7, 2010 06:41
|
|
|
dwoloz posted:I found a clock radio with its power cord yanked in a dumpster that I fancied fixing and using. Turns out the ripped out cord also ripped out a chunk of the PCB.
|
|
#
?
Jun 7, 2010 11:46
|
|
|
http://en.wikipedia.org/wiki/List_of_7400_series_integrated_circuits http://en.wikipedia.org/wiki/List_of_4000_series_integrated_circuits I always use that list when I can't remember what a 7400-series part does. Edit: Huh, the links to ripped-off datasheetcatalog.com datasheets are new. I don't see why they don't link to the ones supplied by a manufacturer instead. Texas Instruments still has them all in their website and they're the ones who invented (I think?) the 7400 series. ante posted:Part numbers can be looked up on something like http://www.datasheetcatalog.com/ I hate that site and others like it. When I do a search for a part number, I want the actual manufacturer's site to come up first, not buried somewhere on the second or third page under a dozen sites like that one. Most of them don't even seem to actually have datasheets anywhere, just a big list of part numbers they probably ripped from a catalog. And if they do have it they make you click through a bunch of crap (for maximum ad exposure!) before they let you see it. And when you finally do get to the datasheet, it has an ad for the website inserted into it. gently caress those leeching assholes. BattleMaster fucked around with this message at 12:23 on Jun 7, 2010 |
|
#
?
Jun 7, 2010 12:15
|
|
|
|
| # ? Jun 10, 2024 23:42 |
|
|
Randel Candygram, i'd be like scrooge mcduck and just swim in all those ICs and then get sent to the ER with acute lead poisoning and puncture wounds. i've got a question, though. Can someone explain to me how an H-bridge works. like in simple terms that i can easily repeat?
|
|
#
?
Jun 7, 2010 15:05
|
|