|
Well the dev kit I'm working with connects via USB, so putting a USB chip on the daughtercard doesn't make much sense. And let me clarify that this is for educational purposes, and that this is a FPGA we're working with, not a MCU. So I'm not looking for external interface chips; the point of the FPGA is that any interfacing logic should be implemented on the FPGA. I just need to know what kind of physical connectors to put there, and, if necessary, what hardware to put in to make it work (logic level conversion for RS232, pulse transformers for ethernet, etc). The exact application of the daughtercard is unclear. Previously, the class used the PS/2 and VGA connector to ultimately design a simple calaculator that interfaced via keyboard and displayed on a vga monitor (making a calculator is easy, but the interfacing is the hard part). With this new card, it's likely that the class will be entirely redesigned around its capabilities. I've only talked with the guy who teaches the course a little bit, and so far I know I'll want an on board H bridge for motor control, a DAC and ADC for basic DSP applications, connectors for rotary encoders, and probably PS/2 and VGA connectors. Until we actually decide what the labs will be, we won't know exactly what we'll need, but I'm trying to at least get some ideas. My main concern right now is having a 50+ watt motor driver, a 12 bit analog sampling system, and a 200MHz FPGA operating together without noise interference... it's a hellish combination of noisy and noise-sensitive systems. edit: Though not as bad as a telemetry board sitting next to the drive circuitry for a 60kW tesla coil driver 
ANIME AKBAR fucked around with this message at 20:55 on Dec 19, 2009 |
#
?
Dec 19, 2009 20:48
|
|
|
|
| # ? May 21, 2024 12:03 |
|
|
Chuck on plenty of ADC channels eh. As far as rotary encoders go, see about having support for both quadrature encoders (with A/B + index channels) plus parallel-connected absolute encoders. You can just use any old SPI port for serial-connected absolute encoders.
|
|
#
?
Dec 20, 2009 01:13
|
|
|
Eh, just giving some ideas, you can hard code the MCU to make it more challenging to interface with. So'kay if not, I totally understand not wanting to mess around with more complexity than you have to. Also, why not move the h-bridge drive MOSFETs/BJTs/IGBTs/ICs to a daughtercard connected by a cable, or even better an el-cheapo fiber link? That would help isolate noise pretty well. Art of Electronics has a $10 (or so) fiber link circuit for just these types of apps, and I'm sure you can find cheaper components nowadays. Sounds like motor control is going to be a big part of that lab, from the shopping list you put up...
|
|
#
?
Dec 20, 2009 04:47
|
|
|
Yeah, I'd say a separate board + supply with isolated inputs (check out the ADuM3100), you'd be crazy to put an H-bridge on the same board as any analogue signals. Same board but isolated supply might be ok too.
|
|
#
?
Dec 20, 2009 05:51
|
|
|
I'd really hate to have to make a daughter card for my daughter card with a separate supply. And fiber is definitely not necessary. I'm sure I'll be able to make a layout that will eliminate most interference. Fortunately if you ever want to use the ADC and the motor driver in the same application, you probably won't care about information while the bridge is switching, so the switching noise won't be a problem at all.
ANIME AKBAR fucked around with this message at 06:22 on Dec 20, 2009 |
|
#
?
Dec 20, 2009 06:20
|
|
|
ANIME AKBAR posted:Fortunately if you ever want to use the ADC and the motor driver in the same application, you probably won't care about information while the bridge is switching, so the switching noise won't be a problem at all.  Just go same board, separate supplies, and isolated logic. And of course a good layout.
|
|
#
?
Dec 20, 2009 06:48
|
|
|
Why isolate digital lines? EMI shouldn't be a problem; the nice thing about H bridges is that you can confine your high di/dt paths to very small loops. As for supply rail interference, the supplies for the motor and the rest of the system won't be isolated, but will be regulated independently, so the only path for interference will be circuit ground. That's really all I'll have to worry about, and as long as I implement a star ground layout, ground bounce shouldn't be a problem.
|
|
#
?
Dec 20, 2009 07:02
|
|
|
Success! Click here for the full 1408x1056 image. Thanks to everyone to helped me out, especially Delta-Wye. Design is a bit different than what I originally stated, but I'm happy with how it turned out. Only thing I wished I could have done is maybe solder everything down, but with 40 LEDs total that have 2 "wires" each, not to mention the jumpers and resistors, it would mean about 100 things to solder. I've only done simple things like solder new capacitors to replace busted ones on old motherboards and I'm not too good at is as of now so I don't trust myself doing it right 100 times; not to mention the prototyping board I have here has holes that are way too loving close together for my experience level. So I ended up just bending all the wires down. My thumb loving hurts so bad. A cool little quirk though - since I'm using a "universal" adapter with variable voltage, you can set it to lower voltages to lower the brightness, albeit in a pretty ghetto way. One last concern. When it's running (on the voltage I designed it for - 12V) it has a slight burning smell. Is this something I should be kinda worried about? Do LEDs need to be "burned in" a bit when they're fresh? Is it because of my piss-poor job at painting with conductive paint? Or is it pretty much negligible? Again, thanks for the help guys!
|
|
#
?
Dec 20, 2009 10:00
|
|
|
Hah, that's pretty neat. But no, no you should not smell burning. Touch your resistors, they shouldn't be hot to the touch. Neither should any of your LEDs.
|
|
#
?
Dec 20, 2009 10:07
|
|
|
The LEDs are definitely warm. Also the jumpers on the top are warm - the wires of them of course. If I could feel it through the rubber that would be pretty alarming.
|
|
#
?
Dec 20, 2009 10:25
|
|
|
The LEDs can be a little warm, but not hot. The jumpers are warm? That's a little hosed up.
|
|
#
?
Dec 20, 2009 10:35
|
|
|
quote:One last concern. When it's running (on the voltage I designed it for - 12V) it has a slight burning smell. Is this something I should be kinda worried about? Do LEDs need to be "burned in" a bit when they're fresh? Is it because of my piss-poor job at painting with conductive paint? Or is it pretty much negligible? Looks cool! I suspect you have something set up wrong on your circuit. If I had to guess based on your symptoms, I would guess your LEDs were in parallel or something, but they appear right in the picture. What size resistors are you using? I'm having trouble making it out. If you have a voltmeter or multimeter I would be curious to see the voltage measurements across vcc<->gnd, anode<->cathode a set of 5 LEDs, and pos<->neg on one of the resistors (I guess you could use two of the three should be able to derive the third, but whatever).
|
|
#
?
Dec 20, 2009 10:44
|
|
|
Alright, here's how I have it set up: Click here for the full 1408x1056 image. I have 8 parallels of 5 LEDs each. There are four 100 Ohm resistors for the red LEDs (2.0 forward voltage) and four 75 Ohm resistors for the green LEDs (2.1 forward voltage). Here are the voltage measurements from my multimeter. "Universal" adapter (500mA max) set to 12V. Vcc-to-gnd: I'm not sure what this is. Never heard of the term Vcc myself. Anode-to-cathode of a set of 5 LEDs: 11.50V Positive-to-negative on one of the resistors: On a 100 Ohm resistor - 3.39V ish On a 75 Ohm resistor - 2.08V ish The readings didn't want to sit still for these, if that's something to consider. For the sake of it, I also took the reading from positive-to-negative from one end of the circuit to the other and got 14.5V. It's set to 12V, but since the adapter is unregulated and very possibly a big piece of poo poo it's more than what I have it set it up for. Perhaps that's the reason? The voltage of the adapter by itself when set to 12V is like 18.5V or something. I don't remember exactly, but it's in the 18V range.
|
|
#
?
Dec 20, 2009 11:35
|
|
|
I made a little solar tracker as a group project, thought some of you guys might be interested:  Click here for the full 1280x960 image.  Click here for the full 1280x960 image.  Click here for the full 1280x960 image. Main thread in the Cavern of Cobol (I mostly worked on the microcontroller) http://forums.somethingawful.com/showthread.php?threadid=3240768
|
|
#
?
Dec 20, 2009 12:52
|
|
|
PilotEvan posted:For the sake of it, I also took the reading from positive-to-negative from one end of the circuit to the other and got 14.5V. It's set to 12V, but since the adapter is unregulated and very possibly a big piece of poo poo it's more than what I have it set it up for. Perhaps that's the reason? The voltage of the adapter by itself when set to 12V is like 18.5V or something. I don't remember exactly, but it's in the 18V range. Vcc is a term for the positive supply voltage. It has to do with the collector pin on transistors yadda yadda. I actually looked up the wiki page to post here, but upon reading it realized it really doesn't matter. You ended up taking the reading anyways, so it's all good. With an unregulated adapter you are usually setting up a voltage divider circuit with your load as one of the legs. If you draw the rated current, everything works out great and you get the expected voltage. If you draw too much current, the voltage sags, and in your case, too little current makes the voltage go up a bit. I wouldn't worry about it, but you should be aware that you are drawing too much current in both legs, most likely. If you look at the voltage across your resistors: 100 ohm: V = IR -> I = V/R = 3.4 / 100 = 34 mA 75 ohm: I = V/R = 2.1 / 75 = 28 mA It will probably run for a while like that, but you can reduce power consumption, the burning smell, and still get enough light if you raise the resistances a bit. I would probably bump the 100 ohm resistors up to 200 ohm or so, and the red to 150 maybe. It's hard to say, because once you have reduced your current draw, the voltage will go up some. I guess you can probably figure out the internal resistance of the adapter pretty easy and that would let you calculate it exactly, but guess and check is much less math intensive  LEDs are fairly resilient though, you can push them like this for a while and they will probably be okay.
|
|
#
?
Dec 20, 2009 20:23
|
|
|
If the jumpers are close to the terminals of the LED, then they might get warm because of thermal conduction. The smell is most likely resistors. LEDs rarely smell unless they've been completely burned out, while resistors can still function while burning. I can't see any thing extraordinary with your voltage measurements (you're not exceeding 1/8 watt on any resistors), but if you can smell it, then something has to change. After all, what you're smelling is the result of a chemical reaction, which means something is being burned away... Just cut the current draw in half or something. Radiance (the actual power emitted from the source) is roughly proportional to current, but luminous intensity (the actual brightness your eye perceives) is roughly proportional to the square root of current. This means that in terms of brightness, LEDs are more efficient at lower average currents. This is good for you because you can cut your power substantially while only slightly decreasing brightness.
|
|
#
?
Dec 20, 2009 21:03
|
|
|
ANIME AKBAR posted:If the jumpers are close to the terminals of the LED, then they might get warm because of thermal conduction. The smell is most likely resistors. LEDs rarely smell unless they've been completely burned out, while resistors can still function while burning. Also I think he used a conductive pen instead of solder.
|
|
#
?
Dec 20, 2009 22:55
|
|
|
I did use a conductive pen and to be honest I did a pretty lovely job with it. The paths are pretty wide and because the pen was so unreliable when I tested it on a different prototyping board, I ended up squeezing the paint on a pallete and using a paintbrush to paint the jumpers. But I'm positive that it's all been painted correctly. I was insanely careful about not crossing parallels when painting. So I doubled the resistance on everything. Here are the differences, all of which Delta-Wye was correct about: -LEDs are definitely a bit dimmer (that luminous intensity thing that ANIME AKBAR mentioned is pretty interesting). With the original resistors I thought it was perhaps too bright, but this maybe seems a bit too dim, but I think I'll keep it for the rest of the differences. -While the LEDs are still a bit warm, which as ante mentioned that should be fine, the jumpers aren't. They feel like they might be very slightly, but I can't really tell well enough to make a clear statement between not warm and warm. So that's good. -The burning smell is also diminished. Also good. I'm sort of hoping the burning smell is from the conductive paint, and that it's burning off the rough parts or something. I honestly wouldn't know how that works so that's what I've pulled out of my rear end. Should I keep it running for a while to see if it stops after a while or not? PilotEvan fucked around with this message at 23:59 on Dec 20, 2009 |
|
#
?
Dec 20, 2009 23:46
|
|
|
PilotEvan posted:I'm sort of hoping the burning smell is from the conductive paint, and that it's burning off the rough parts or something. I honestly wouldn't know how that works so that's what I've pulled out of my rear end. Should I keep it running for a while to see if it stops after a while or not? Solder that poo poo hombre.
|
|
#
?
Dec 21, 2009 00:07
|
|
|
Well I've been running it for the past few hours and the burning smell is gone. I didn't do any soldering but I don't think I need to anymore. Looks like it's set to go! Thanks guys!
|
|
#
?
Dec 21, 2009 02:25
|
|
|
Just picked up a 350 MHz 4-channel Tektronix scope off eBay. Merry Christmas Cyril Sneer! 
|
|
#
?
Dec 21, 2009 05:00
|
|
|
I need a decent beginner's soldering station, is the Weller WLC100 good enough for starting (I will be working with microcontrollers) or should I spring for something better?
|
|
#
?
Dec 21, 2009 17:00
|
|
|
Metajo Cum Dumpster posted:I need a decent beginner's soldering station, is the Weller WLC100 good enough for starting (I will be working with microcontrollers) or should I spring for something better? That's the exact model I have, and it works fine for a soldering iron. You may need to move to a hot air system or something eventually, but its acceptable for what it is.
|
|
#
?
Dec 21, 2009 18:20
|
|
|
Delta-Wye posted:That's the exact model I have, and it works fine for a soldering iron. You may need to move to a hot air system or something eventually, but its acceptable for what it is. Thanks. What's a hot air system? Also, my apartment is fairly small and enclosed, and the place I plan on having it is not near a window. Will the flux fumes be a problem?
|
|
#
?
Dec 21, 2009 18:25
|
|
|
Metajo Cum Dumpster posted:Thanks. What's a hot air system? Dude, the flux fumes are the best part of soldering Something like this would be great if you were totally serious and would grow into it: http://www.sparkfun.com/commerce/product_info.php?products_id=76 I have not regretted buying that Weller at all, but I have considered replacing it with something like that lately.
|
|
#
?
Dec 21, 2009 20:22
|
|
|
Hot air is best for rework - taking off parts once they have been soldered. It makes it trivial to unsolder an entire IC, even ones with hundreds of pins. If you are mostly building boards from scratch, all you need is an iron.
|
|
#
?
Dec 21, 2009 20:39
|
|
|
Delta-Wye posted:Dude, the flux fumes are the best part of soldering I have one of these rework stations (well there are a lot of types like that which are essentially identical) and it's pretty shoddy. I don't trust the temp control on the soldering iron at all, and the cord is really short. The hot air gun is okay, but I think my unit is broken because whenever I turn the airflow up to a certain point it makes nasty noises... Anyways for a soldering iron, I'd go with a weller to start with, and a metcal smartheat system if you're looking for something nice. Ebay is a great place to look. As for hot air systems, I really haven't seen much besides those cheap korean ones.
|
|
#
?
Dec 23, 2009 17:23
|
|
|
Thanks. Also, I need a slightly better digital multimeter than the $10 one I have. I'm not talking $300 Fluke better, but what about these two Mastech ones? Manual http://www.amazon.com/Mastech-Multimeter-Capacitance-Measurement-MS8261/dp/B000FLZDPG/ Auto-range (I don't know if auto-ranging is good or bad?) http://www.amazon.com/Mastech-Manual-Digital-Multimeter-MS8268/dp/B000JQ4O2U/
|
|
#
?
Dec 23, 2009 17:52
|
|
|
I just checked, and I use a Weller WCC 100 iron and a Hakko 850 hot air rework station in my office at work. We have other stuff in the lab too. Temp control isn't a big deal on the irons. I just use them all on max (800-850° or so) and have never burned out a part.
|
|
#
?
Dec 23, 2009 18:49
|
|
|
Oh yeah and the ones that say they are ESD safe are straight up lying. I've noticed that LEDs light up when I do rework on unpowered boards, meaning the tips aren't even isolated. this is the model I bought, but you can tell that all of them are built with the same exact parts and are only cosmetically different.
|
|
#
?
Dec 23, 2009 20:38
|
|
|
... it is kind of never break...
|
|
#
?
Dec 23, 2009 20:43
|
|
|
oh yeah, it's obviously sketchy as hell, but so is everything else out there. I really don't know any reputable companies that make affordable hot air stations. OK and Cooper make some, but they're drat expensive.
|
|
#
?
Dec 23, 2009 20:56
|
|
|
Help. I'm retarded and new to electronics. I also can't interpret circuits into breadboard layouts worth a poo poo. I tried this circuit, and the Vout is actually 1/2Vin  1: For the Vin I want to use 2x1.5v batteries, which of the Vin dots is (+)? 2: Is a 1N914 okay to use for the Dx? What capacitance value should I use for the caps? 3: Diode diagrams confuse me; So the direction of the diode shows the flow of "holes" and is reverse of the direction of the flow of electrons?
|
|
#
?
Dec 24, 2009 05:13
|
|
|
I've been using my $20 Radioshack adjustable power soldering iron for years now. Why would it be advantageous to get a workstation?Metajo Cum Dumpster posted:Help. I'm retarded and new to electronics. I also can't interpret circuits into breadboard layouts worth a poo poo. 1. Top is always positive. 2. I'm on break from shcool, so I'm not thinking about circuit analysis very well right now.(Doesn't look like it would work, but I'm probably wrong) 3. The "flow of holes" school of thought never made any sense to me. The arrow in the diode symbol points in the direction of the flow of current. In real life, the band on the diode symbol matches up with the band on the component, directionwise.
|
|
#
?
Dec 24, 2009 06:06
|
|
|
Metajo Cum Dumpster posted:Help. I'm retarded and new to electronics. I also can't interpret circuits into breadboard layouts worth a poo poo. that circuit is a rectfier: if you put in an AC voltage, the top diode will rectify the positive part of the AC and the lower one will rectify the negative portion of the AC. The voltage between the two outputs will a DC voltage of twice the amplitude of the AC input (or equal to its peak to peak voltage). This circuit won't really do anything with DC voltages.
|
|
#
?
Dec 24, 2009 06:37
|
|
|
ANIME AKBAR posted:that circuit is a rectfier: if you put in an AC voltage, the top diode will rectify the positive part of the AC and the lower one will rectify the negative portion of the AC. The voltage between the two outputs will a DC voltage of twice the amplitude of the AC input (or equal to its peak to peak voltage). This circuit won't really do anything with DC voltages. ante posted:I've been using my $20 Radioshack adjustable power soldering iron for years now. Why would it be advantageous to get a workstation? I badly want someone to convince me to spend large amounts of money on more poo poo I don't need 
|
|
#
?
Dec 24, 2009 07:00
|
|
|
ante posted:I badly want someone to convince me to spend large amounts of money on more poo poo I don't need Fast to come up to temperature (really really fast), safer (auto shutoff), more accurate temperatures, better more flexible cables, easier to find and change out tips to different sizes. Seriously, it'll blow your mind. Even just one of the $100 Weller WES series is awesome. In related news, I'm building a Nixie clock after years of wanting one. 
|
|
#
?
Dec 26, 2009 05:18
|
|
|
ante posted:I badly want someone to convince me to spend large amounts of money on more poo poo I don't need there's no reason to spend a lot of money, though. Check ebay, or better yet, find an industrial surplus warehouse. You can find stuff like a loving box of weller soldering stations for $30 each. Many surplus places will even ship you stuff if you aren't local.
|
|
#
?
Dec 28, 2009 20:32
|
|
|
NM.
Canary Yellow fucked around with this message at 15:00 on Dec 30, 2009 |
|
#
?
Dec 30, 2009 13:31
|
|
|
|
| # ? May 21, 2024 12:03 |
|
|
Thanks for the info on the circuit. Another question; My old rear end samsung i730 smartphone died and I pulled the touchscreen. Is there any way to use it with the right controller or is it device specific or whatever?
|
|
#
?
Dec 31, 2009 18:18
|
|