|
The only correct solution here is to a finite element analysis of this general problem, store the results in a predefined lookup table, and then do a calibration to correct for the differing resistivity of a new material or a change in the size of the square. Yeah, it's not trivial, but there's a reason why resistive touchscreens aren't implemented with a single resistive layer. sixide posted:I think the right way to do this is a mathematical transform. It's pretty clearly a simple (conic?) relationship, and given that you can assume the surface is uniformly resistive it should be fairly straightforward to figure it out even from first principles. The problem with this is that any generalized transform would get hosed up to some extent by the distortion near the edges (depending on how near the electrodes are to the edge of the board). Slanderer fucked around with this message at 23:13 on Aug 24, 2013 |
#
?
Aug 24, 2013 23:10
|
|
|
|
| # ? Jun 3, 2024 23:05 |
|
|
Slanderer posted:The only correct solution here is to a finite element analysis of this general problem, store the results in a predefined lookup table, and then do a calibration to correct for the differing resistivity of a new material or a change in the size of the square. Resistivity and square size are completely irrelevant to the problem. A LUT is probably the easiest in terms of up-front math, but it's complete poo poo in processing speed and logic burden. "Distortion" really isn't the right name for this effect. There's a very clear (x,y) = F(x voltage, y voltage) relationship that isn't some crazy piecewise empirical function here. https://docs.google.com/spreadsheet/ccc?key=0AvZK5TSYzUa0dEdhZXItQmdad2poSTM1NTAxVTNVdUE&usp=sharing Voltage (x-mode) for the uniform square resistive sheet. Usable with a LUT, but I can't quite figure out what an appropriate curve for constant x vs. y is. Not parabola, not catenary, not ellipse, not hyperbola, not several other things. One of those times I wish my analysis skills weren't complete poo poo.
|
|
#
?
Aug 27, 2013 01:14
|
|
|
Sorry, I wanted to dig into these responses by now, but I'm going to be out of town for a week, so I'll catch up when I'm back. Mentioning this so if someone drops an awesome solution to the distortion problem you don't think I ran off without honoring my  offer.
|
|
#
?
Aug 28, 2013 01:17
|
|
|
What kind of software would I need to test out a CMOS logic circuit? I took a final exam last semester and one of the questions involved designing a CMOS circuit for a given logic function and I was marked wrong (this one question means the difference between an A and a B in the course so I am invested in proving my answer) and I want to verify my circuit. I am hesitant to reinstall MultiSim and try it with that because I remember not having a MOSFET available in my version of MultiSim and having to model it with a voltage regulator or something.
|
|
#
?
Aug 28, 2013 21:37
|
|
|
IratelyBlank posted:What kind of software would I need to test out a CMOS logic circuit? I took a final exam last semester and one of the questions involved designing a CMOS circuit for a given logic function and I was marked wrong (this one question means the difference between an A and a B in the course so I am invested in proving my answer) and I want to verify my circuit. I am hesitant to reinstall MultiSim and try it with that because I remember not having a MOSFET available in my version of MultiSim and having to model it with a voltage regulator or something. LT Spice is the goto free simulation tool. Or post a pic here and I'm sure people will analyze it for you.
|
|
#
?
Aug 28, 2013 23:59
|
|
|
I'm working on making an Arduino LED strip music visualizer thing, and I'm running into some noise from PWMing the LED strip through the MOSFETs. The LED strip is an "analog" 5050 12v strip, which I'm using MOSFETS with the Arduino to control. I get an audible hum over my speakers when the LEDs are powered, and occasionally will get a feedback loop that will keep one of the LEDs lit, causing more feedback which puts noise on the MSGEQ7 input, etc. I'm wondering what I can do to eliminate this. When I separate the 12v and 5v circuits and power them with separate power supplies, the noise goes away, but I'm looking to have this powered by a 12v battery, so I can't do this in my intended application. Here's the guide I used for wiring the LED strips: http://learn.adafruit.com/rgb-led-strips Here's the guide I used for wiring the MSGEQ7: http://nuewire.com/info-archive/msgeq7-by-j-skoba/ This is the Arduino on the breadboard: http://www.instructables.com/id/YABBAS-Yet-Another-Bare-Bones-Arduino-on-Stripb/?ALLSTEPS Here are a couple pictures of my breadboard, with and without notes: http://i.imgur.com/IzmhGcT.jpg http://i.imgur.com/OzGFeDV.jpg A video of it working is here: http://www.youtube.com/watch?v=RtVW7ehgaK4 Thanks!
|
|
#
?
Aug 30, 2013 02:57
|
|
|
Sounds like you're coupling switching noise back onto the analog audio. It may help to adjust the PWM frequency out of the audible range, the noise may just totally disappear: http://playground.arduino.cc/Main/TimerPWMCheatsheet Otherwise, you need to clean up the power filtering. More bypass capacitors may help, connecting the audio ground through a choke may help, I'm really not sure what I would do. Now that I don't have access to nice oscopes to try and characterize the noise, I would probably just start trying one thing after another. If it is switching noise, it would probably sound like a garbled pitch, roughly middle B. I guess if you have an ear for music that may help?
|
|
#
?
Aug 30, 2013 06:54
|
|
|
^^changing the frequency is an excellent idea. Bring the power onto the breadboard with as short/fat wires as possible and bypass it well. 20uf minimum, 100+ would be better. Tap your audio/control power off immediately and bypass further at every IC with 1uf caps. Take two wires and create an LED PWM section on the opposite side of the board (an inductor or ferrite between this and the control/audio section would be ideal). Bypass this local power island further and locate the mosfets here. Add a resistor in series with the gate of the mofets, I'd say 1k if your PWM frequency is 50k or less (otherwise I'd double check this value). This is important because it will slow down the turn on/offs and cut down switching noise. Also consider a regulator as a way isolate the power (or do you have one already?) Off the top of my head that's what I would do.
|
|
#
?
Aug 30, 2013 17:17
|
|
|
It's definitely related to the PWM switching. I've got a few other modes other than the audio reactive programmed in, which I can switch between using the buttons, and use the pot to adjust various aspects of, and the sound of the noise through the speakers changes as the PWM duty cycle changes. I've tried putting a bigger capacitor on the 12v supply, but that didn't seem to help. I'll play around with changing the PWM frequency tonight. That may be good enough to fix my problem since the MSGEQ7 chip only responds to audible frequencies anyhow. If I do try a choke, is there any particular part or value I should look for? asdf32 posted:Bring the power onto the breadboard with as short/fat wires as possible and bypass it well. 20uf minimum, 100+ would be better. Tap your audio/control power off immediately and bypass further at every IC with 1uf caps. Yeah, there's a 7805 power regulator on the arduino board that takes the 12v required by the LED strip and puts out 5v that everything else is using. I've already got the source and drain of the mosfets and power to the 12v strip on its own section of the board, with just the PWM outputs from the arduino connected to the gate. I'll try the resistors as well, if the frequency change doesn't help. I'm also going move more toward a "star" ground scheme on the 5v side as well, I've already done so with the 12v ground from the mosfets. Thanks again for the help. n0tqu1tesane fucked around with this message at 17:29 on Aug 30, 2013 |
|
#
?
Aug 30, 2013 17:21
|
|
|
http://makezine.com/projects/Sous-Vide-Immersion-Cooker/ I am going to be making this over my vacation but am wondering how hard it would be for someone who has ZERO knowledge or experience to change the plan to use plugs for the various parts so that it is more modular. I have heard that the immersion heaters have a tendency to break and I don't want to have to rewire things after the fact if it does. It is far nicer to just plug the new part in.
|
|
#
?
Aug 30, 2013 18:17
|
|
|
demonachizer posted:http://makezine.com/projects/Sous-Vide-Immersion-Cooker/ Buy some cheap extension cords and cut the ends that you need off and wire those directly instead of the immersion heaters.
|
|
#
?
Aug 30, 2013 18:31
|
|
|
n0tqu1tesane posted:Buy some cheap extension cords and cut the ends that you need off and wire those directly instead of the immersion heaters. Yeah I was just thinking that. That is probably the easiest and won't require me to depart from the parts list much. I hope that I don't kill myself but I figure it is a good way to determine if I am retarded or not.
|
|
#
?
Aug 30, 2013 18:35
|
|
|
Yeah. If you really wanted to go all out with it you could wire them into some GFCI outlets for added protection in case the whole thing plops into your cooking bucket.
|
|
#
?
Aug 30, 2013 19:51
|
|
|
Anyone do anything with PWM on the MSP430s? Specifically the FRAM 5969 model? The resolutions seem suspiciously low to me...is it possible / usable to chain the CCRx registers to get an effective 32-bit timer?
|
|
#
?
Aug 31, 2013 00:28
|
|
|
Delta-Wye posted:Sounds like you're coupling switching noise back onto the analog audio. It may help to adjust the PWM frequency out of the audible range, the noise may just totally disappear: Changing the PWM frequencies worked! Thanks for your help. EDIT: Still getting a very occasional pop and hiss, but it's so quiet that I doubt it'll be noticeable in a room that's not completely silent, or to a person who's not listening for it with their ear next to the speaker. n0tqu1tesane fucked around with this message at 01:42 on Aug 31, 2013 |
|
#
?
Aug 31, 2013 01:34
|
|
|
demonachizer posted:http://makezine.com/projects/Sous-Vide-Immersion-Cooker/ If you have literally zero experience, try this: http://learn.adafruit.com/sous-vide-powered-by-arduino-the-sous-viduino/sous-vide You don't get the aquarium pump water circulator (though you can easily add one), but the rice cooker has much better insulation than that plastic container he's using. The design is also actually more modular than it looks because the complicated part, the PID controller, is all separated from the heater and vessel, so if you want to upgrade later to something bigger it's quite straightforward.
|
|
#
?
Aug 31, 2013 02:44
|
|
|
n0tqu1tesane posted:Changing the PWM frequencies worked! Thanks for your help. Here's a video of everything working: https://www.youtube.com/watch?v=fleFgcl8dNo Now, I've just got to figure out what I'm going to do for an enclosure, move everything to a PCB, and get it installed in my coworkers rolling sound system.
|
|
#
?
Aug 31, 2013 18:12
|
|
|
I want to protect some circuitry from accidental contact with other energized components. Can I just cover the thing with construction silicone? The can says "non-acidic" but I am not sure whether that means it is safe to use with electronics. How can I tell?
|
|
#
?
Sep 2, 2013 13:50
|
|
|
EssOEss posted:I want to protect some circuitry from accidental contact with other energized components. Can I just cover the thing with construction silicone? The can says "non-acidic" but I am not sure whether that means it is safe to use with electronics. How can I tell? If it smells like vinegar, don't use it. Look at potting compounds specifically, but for something consumer-grade I bet an off-the-shelf silicone would work well enough.
|
|
#
?
Sep 2, 2013 13:58
|
|
|
EssOEss posted:I want to protect some circuitry from accidental contact with other energized components. Can I just cover the thing with construction silicone? The can says "non-acidic" but I am not sure whether that means it is safe to use with electronics. How can I tell?
|
|
#
?
Sep 2, 2013 14:14
|
|
|
Home depot sells spray on electrical tape for 8 bucks a can.
|
|
#
?
Sep 2, 2013 23:17
|
|
|
For my hobbyist stuff I use a tube of GE Silicone II. It doesn't release acetic acid while curing, and has worked for me thus far.
|
|
#
?
Sep 3, 2013 21:38
|
|
|
If either of the parts you want protected are small enough, heat shrink is always great to use.
|
|
#
?
Sep 4, 2013 04:07
|
|
|
Terminal Entropy posted:If either of the parts you want protected are small enough, heat shrink is always great to use. Heat shrink tubing makes my stuff look so pro if I trim it nicely. My question: I have some 62.5V batteries which are old, but I think they still have quite a few mA-hours left to give this world. If I want to use them for 5V electronics, I assume a 7805 regulator is going to waste a tremendous amount of energy. And I assume putting a large current limiting resistor in series is not going to play nicely with the 7805 because the load's variable current draw will cause the voltage to fluctuate. Are my only options acquiring 62.5V-capable 5V switching regulators?
|
|
#
?
Sep 4, 2013 20:20
|
|
|
EssOEss posted:I want to protect some circuitry from accidental contact with other energized components. Can I just cover the thing with construction silicone? The can says "non-acidic" but I am not sure whether that means it is safe to use with electronics. How can I tell? Hot melt glue also works.
|
|
#
?
Sep 4, 2013 21:00
|
|
|
reading posted:Heat shrink tubing makes my stuff look so pro if I trim it nicely.  62.5 volts into a 7805? Don't put it on any flammable surfaces!Yes, your best option here is an appropriate switching regulator. You aren't going to get a whole lot of efficiency stepping the voltage down that much, but at least you won't be burning up 93% of the energy as heat and instantly detonating your poor 7805. Out of curiosity, what on earth uses a 62.5 volt battery? The only things I can think of are those really old portable ham radios and the like that needed the high voltage before boost converters or charge pumps were invented.
|
|
#
?
Sep 4, 2013 21:23
|
|
|
What's the cheapest way to drive a couple 800mA brushless motors? I've got a couple L293Ds in there now, but I need to be able to drive the enables seperately, and at $7 a pop, there's gotta be a better way. $21 driver circuitry is pretty not cool. That's not even including a microcontroller with 6 outputs + 6 PWM + 3 inputs. I'm not above using a stack of twenty cent MOSFETs or whatever, if necessary.
|
|
#
?
Sep 4, 2013 21:51
|
|
|
Here's a $6 driver circuit that only needs the one wire: http://www.hobbyking.com/hobbyking/store/__18002__HobbyKing_Red_Brick_10A_ESC.html
|
|
#
?
Sep 4, 2013 21:57
|
|
|
ante posted:What's the cheapest way to drive a couple 800mA brushless motors? I used a dsPIC33F + BD6221Fs to drive 3 motors in either direction, wasn't horribly complicated.
|
|
#
?
Sep 4, 2013 22:22
|
|
|
Sagebrush posted:Here's a $6 driver circuit that only needs the one wire: http://www.hobbyking.com/hobbyking/store/__18002__HobbyKing_Red_Brick_10A_ESC.html I was hoping to have some sort of fine control of the motors. They have mounted Hall Effect sensors and everything. This is a cool solution for something else I have planned for these motors, though. Any idea what chips they're using? movax posted:I used a dsPIC33F + BD6221Fs to drive 3 motors in either direction, wasn't horribly complicated. Something like that chip would be awesome, but these are three-phase brushless DC motors. I need three half-bridges instead of an h-bridge.
|
|
#
?
Sep 5, 2013 00:18
|
|
|
ante posted:I was hoping to have some sort of fine control of the motors. They have mounted Hall Effect sensors and everything. This is a cool solution for something else I have planned for these motors, though. Any idea what chips they're using? These ESCs use either a cheap AVR controller or maybe one from SiLabs. Many of them, from different manufacturers, are nearly identical, which lead to the creation of an improved custom firmware for a lot of them: https://github.com/sim-/tgy
|
|
#
?
Sep 5, 2013 00:58
|
|
|
Cross posted from: http://forums.somethingawful.com/showthread.php?threadid=3568456 Howdy fellow goons. A friend of mine (the girl in the video) has been working on this for some time now and I thought Serious Hardware would find this to be Seriously Cool. http://www.crowdsupply.com/octopart/octopart-pocket-electronics-reference-pcb It's a really well designed resistance cheat sheet + footprint size chart. She wants to get some feedback directly from the goons. What do you guys think? Here's a pic of one of the boards (without the silk screening) 
|
|
#
?
Sep 5, 2013 01:23
|
|
|
Fingertips posted:Cross posted from: http://forums.somethingawful.com/showthread.php?threadid=3568456 I already spent all my electronics crowdfunding moneys on the µRuler but that thing does indeed look sharp. My current resistor color chart is just a business card that came with an order I placed a long time ago so it would be nice to have something more permanent.
|
|
#
?
Sep 5, 2013 01:33
|
|
|
Fingertips posted:Cross posted from: http://forums.somethingawful.com/showthread.php?threadid=3568456 The color code isn't that useful for me, but the footprints on the back are kind of neat. Still, should be good for hobbyists and I'm really interested in checking out how well all those different silkscreen colors turn-out post-production!
|
|
#
?
Sep 5, 2013 08:17
|
|
|
Router? I hardly even know her   My girlfriend is getting a lamp stuffed with CFLs and LEDs for her birthday. The plan is to prototype on Arduino, and then burn the final program on an atmega168 that is mounted on the board seen above. I am pretty much learning as I go here; never made a PCB before, never built my own arduino hardware, never written a sketch of this magnitude... The electronics are nothing too fancy, just some voltage regulation and MOSFETs, most the magic is software-side. The software is about 70% done as well. It takes commands from an IR remote, and most of the work now is programming functions for all the individual buttons. Video: https://www.youtube.com/watch?v=64dQuqM_OsM The fading is a lot smoother than the video makes it look like. snorch fucked around with this message at 12:35 on Sep 5, 2013 |
|
#
?
Sep 5, 2013 12:33
|
|
|
reading posted:Heat shrink tubing makes my stuff look so pro if I trim it nicely. Yeah, that is the best thing about heat shrink  
|
|
#
?
Sep 5, 2013 17:14
|
|
|
snorch posted:Router? I hardly even know her I  old-school pcb layouts. The curves scream 'retro', like a copper egg chair or something.
|
|
#
?
Sep 9, 2013 17:03
|
|
|
Terminal Entropy posted:Yeah, that is the best thing about heat shrink Heat shrink is great but I find it difficult to personally stock the various sizes and the high shrink ratio stuff is expensive. Self fusing tape is a great alternative, although it does lack the professional look.
|
|
#
?
Sep 10, 2013 02:39
|
|
|
Harbor Fright has a nice set of assorted sizes that have been precut: http://www.harborfreight.com/127-piece-heat-shrink-tubing-set-67524.html The box it all comes in is nice for general storage too.
|
|
#
?
Sep 10, 2013 02:59
|
|
|
|
| # ? Jun 3, 2024 23:05 |
|
|
If you want to look Monster legit, put a mesh sleeve on your wires, then put transparent heat shrink on that!  
|
|
#
?
Sep 10, 2013 03:04
|
|
