|
Kire posted:I've been playing with my launchpad and doing things like blinking the LEDs on the board, using interrupts with the buttons, controlling a 7-segment LED, and so forth. I hit a bit of a wall when I wanted some PWM output to control an RGB LED (and just cheated by using an arduino since the Atmel chip had hardware PWM). I'd like to get better at mastering the MSP430 but I'm sort of out of ideas, beyond "push a button, ring a buzzer, blink an LED". Without a serial output it's hard to check things like the internal temperature sensor or the ADCs. Not sure, but I use this resource (I fear you might be more advanced than me and those tutorials but worth posting for others too): http://processors.wiki.ti.com/index.php/MSP430_LaunchPad_%28MSP-EXP430G2%29
|
#
?
Feb 4, 2013 13:23
|
|
|
|
| # ? May 31, 2024 10:31 |
|
|
Parallel Paraplegic posted:I added the 1K resistors and no luck. The forward-biasing makes sense, but it's weird that the voltage remains exactly and perfectly unchanged as far as I can tell (down to three decimal places, according to my cheap and unreliable multimeter). I assume that means I can't actually build this circuit with BJT's then? One of the fun and amazing things about the various families of logic is that while their outputs were eventually equivalent, their topologies were not so much. There's very good technological reasons why we went from DL to DTL to RTL to TTL to ETL to CMOS logic, and why the fundamental circuits of that logic shifted around. The "simplest" gate in TTL was the AND, but for CMOS it was NOR. The easiest way to get an AND in CMOS was to build a CMOS NOR then invert it, not to duplicate the TTL AND transistor hookup.
|
|
#
?
Feb 4, 2013 13:43
|
|
|
Parallel Paraplegic posted:Okay, here's what I drew up, including the values of resistors I currently have in there while loving around with it. The problem is that you have the bases of the n's and the p's tied together. That means they're both fighting each other (consider their internal diodes) and they're both turning on. Move those 220s after the switch and put them in series with each of the 4 bases and it will work. Try to always have a resistor in series with the base and you'll be better off (it's not a bad idea for a MOSFET gate either). Edit: whoops responded to quick. I see someone said this already. I'll respond later if I have a chance but the bottom line is that it should work no problem. If you're still burning parts you've got a midwife somewhere because that 100 and diode would probably save you. asdf32 fucked around with this message at 20:34 on Feb 4, 2013 |
|
#
?
Feb 4, 2013 16:41
|
|
|
Parallel Paraplegic posted:I added the 1K resistors and no luck. The forward-biasing makes sense, but it's weird that the voltage remains exactly and perfectly unchanged as far as I can tell (down to three decimal places, according to my cheap and unreliable multimeter). I assume that means I can't actually build this circuit with BJT's then? Here is some information on how a standard TTL NAND gate is arranged: http://www.psut.edu.jo/sites/alasali/Digital%20Electronics%20Text/P-TTL.pdf And a video of one being tested in Falstad: https://www.youtube.com/watch?v=IQGO4YrrjGo The multiple-emitter transistor is built via multiple transistors with the bases and collectors in parallel. Delta-Wye fucked around with this message at 02:34 on Feb 5, 2013 |
|
#
?
Feb 4, 2013 19:00
|
|
|
Parallel Paraplegic posted:Okay, here's what I drew up, including the values of resistors I currently have in there while loving around with it. Ok so based on a second look and reading everything you said your problem still isn't entirely clear. But here are some more points: -Definitely have resistors directly on the bases 1k. Ditch the 220 on the switch, that just confuses things. Have the switch tie directly to GND or 5V. -You don't want the diode and the resistor on the emitter of the NPN's. This complicates and confuses the circuit because the NPN's turn on when current (dependent on voltage) flows between the base and the emitter. If the emitter voltage is floating on top of an R and D, and not tied to ground, it's behavior becomes more complicated - as current increases through the diode the emitter voltage increases across the resistor, this can choke off the emitter current if it gets too high (as an aside you can make a decent current source this way, but that's not what you're doing). Imagine driving a car where the gas pedal moves as the car changes speed - that's what you've got there. -I'd suggest adding an ON and OFF diode, one going from OUT to GND and one going from 5V to OUT. This will be a clean way to see what's happening. -Last, try downloading LT Spice if you're doing this stuff. Here is a picture from LT spice. The voltage sources on the left are the switches, the resistors on the right provide a load on the output: 
asdf32 fucked around with this message at 02:29 on Feb 5, 2013 |
|
#
?
Feb 5, 2013 02:18
|
|
|
asdf32 posted:Ok so based on a second look and reading everything you said your problem still isn't entirely clear. But here are some more points: I'll see if this works, but wouldn't removing the resistor between the NPN's and ground mean that current can flow essentially uninterrupted through the whole thing (effectively bringing back the catching-on-fire issue from earlier?) Should I put a resistor before the PNP's instead? Anyway thanks for everyone's help with this, I never realized the differences between transistor types could make this big a difference. Also, I tried LT Spice a while ago and couldn't get it working (I'm one of those freaks who uses linux, so entirely my fault  ), so I'm actually using what I first learned this poo poo on back in the 90's, CircuitMaker (which works perfectly in WINE for some goddamn reason). It's got full SPICE simulation and does most of the things more modern software does, while being an interface I'm already familiar with, so meh.
|
|
#
?
Feb 5, 2013 03:29
|
|
|
Parallel Paraplegic posted:I'll see if this works, but wouldn't removing the resistor between the NPN's and ground mean that current can flow essentially uninterrupted through the whole thing (effectively bringing back the catching-on-fire issue from earlier?) Should I put a resistor before the PNP's instead? Anyway thanks for everyone's help with this, I never realized the differences between transistor types could make this big a difference. No - If you put resitance at the top of the circuit, between the P's and 5V, then you're exactly replicating the problem I was describing on the N side. The whole point to the topology of the circuit is that you can't have both N's and P's conducting. It takes two 1's to get the N's to conduct at which point the P's are off - when it's set up properly. I suspect that because you originally had the bases tied together that caused the fire and you led yourself in the wrong direction after that. The transistor types don't make that big a difference. In general, a BJT with a resistor on it's base pretty much does the same thing as an mosfet - it's off, then it's kind of on, then it's all the way on as you increase/decrease the gate/base voltage. Also I'll note that if all you want is to drive that out LED when you have either input low then you don't need the N's at all, just delete them. The N's pull the output hard low when you have two ones. If you don't need that hard low then you don't need the N's.
|
|
#
?
Feb 5, 2013 04:53
|
|
|
Humphreys posted:Not sure, but I use this resource (I fear you might be more advanced than me and those tutorials but worth posting for others too): Thank you, this looks great. The PWM tutorial should be helpful.
|
|
#
?
Feb 5, 2013 06:09
|
|
|
Part question: does anybody have a line on where I could pick up a halfway decent inductive pickup for not-assraping prices? I think the lowest I've seen is $80, and that's a little pricey for what I'm going for here. An inductive pickup can't be that hard to make, clamp-on or no. I remember enough of Physics for Wusses II to know that a ferrous bead with the right wire wrapping and a big fat resistor down the line would technically do the trick, but I'd like to avoid doing something you'd expect to see on auto section of thereifixedit.com if at all possible. From what I've gathered, there should be around a 10-15Kv pulse going down the plug wire, and my circuit should only take a couple volts to trip the 555 and send the counter pulse to the processor. As an aside, I went ahead and dropped on an Arduino Uno since they're so simple and newbie-friendly to use, and have a bunch of input protection built in to them in case I do a dumb thing. Once I get this running and verify things are accurate and not shooting crazy input voltages all over the place, I think I'll branch into using a PIC directly instead. Then make a proper garage-tool version that can have its circuit plan minimized and fit into a sealable project box and run off of a 9-volt. Assuming this all works out as intended I'm going to make a few for my dad to use on his ultralight, my uncle to use on his bikes(if he doesn't have one already, the man is the textbook definition of gearhead), and me on own my bike so I don't have to worry about messing up my arduino from dripping oil or dropping a tool on it.
|
|
#
?
Feb 5, 2013 06:30
|
|
|
asdf32 posted:Also I'll note that if all you want is to drive that out LED when you have either input low then you don't need the N's at all, just delete them. The N's pull the output hard low when you have two ones. If you don't need that hard low then you don't need the N's. Really this is more of a "CMOS gates are neato, I wonder if I can build a similar one with the stuff I have laying around" thing, not something I'll probably wind up using outside of playing around with it. The point is more to replicate the original diagram I posted in some form or another rather than needing this to actually do something in another project. Thanks for the help though
|
|
#
?
Feb 5, 2013 06:43
|
|
|
Parallel Paraplegic posted:Really this is more of a "CMOS gates are neato, I wonder if I can build a similar one with the stuff I have laying around" thing, not something I'll probably wind up using outside of playing around with it. The point is more to replicate the original diagram I posted in some form or another rather than needing this to actually do something in another project. Thanks for the help though Actually, it's not hard to build RTL, DTL, or TTL gates with discrete components (in the latter case, making most outputs open collector simplifies things and with bipolar logic you generally pull LEDs down from Vcc anyway). Somebody made a kit for a digital clock with discrete logic...
|
|
#
?
Feb 5, 2013 07:07
|
|
|
I recently upgraded my $20 crummy rear end soldering iron to the Hakko-FX950 soldering station. So - so much better. Can't believe I went through at least 3 or 4 cheapo irons before deciding to spring on a station. It really makes life easier. Now this station has a sleep function with a add-on holder that I haven't got. It connects to the station using a 3,5mm Stereo Jack. That I know by plugging in a standard Stereo cable and measuring the segments with my multimeter. Now, I'd like to use the sleep function via toggle switch, so that I can "sleep" the iron while looking up data sheets etc. Thus -- I'd like to ask if anyone has the toggle holder and if you could take some measurements how its conductive behavior changes (toggled VS un-toggled). I've googled and googled, finding some other mods and hacks but not this. Thanks!
|
|
#
?
Feb 5, 2013 15:40
|
|
|
I'm designing a quickshifter to use on my motorcyle. The basic concept is, when the switch is closed a relay will be used to cut the signal to the ignition coils for 75ms while the gear change is made. It cannot be activated again for ~.5seconds. Does this circuit seem like a good solution?
Zool fucked around with this message at 07:51 on Feb 8, 2013 |
|
#
?
Feb 8, 2013 07:38
|
|
|
My beaglebone has stopped being recognized by my WinXP machine- when I plug in the USB cable, Windows says something like "This USB device is malfunctioning" and it shows up as "Unknown Device" in my hardware list. I usually don't trust Windows to correctly diagnose anything but I cannot get my beaglebone fired up. It powers up ok, the heartbeat light pulses, everything seems normal but between last night and this morning I just can't get it to show up in Windows. Did I kill it with static somehow?
|
|
#
?
Feb 9, 2013 00:37
|
|
|
Zool posted:I'm designing a quickshifter to use on my motorcyle. The basic concept is, when the switch is closed a relay will be used to cut the signal to the ignition coils for 75ms while the gear change is made. It cannot be activated again for ~.5seconds. Does this circuit seem like a good solution? Your link loaded the default circuit simulator circuit, but that might just be my browser loving up the link. Anyway I'm wondering, wouldn't you need some kind of interlock? What if the shift doesn't complete in the 75ms, or the mechanism fails - the engine kicks back on at full power and fucks up poo poo? I don't really know much about motorcycles but it seems like you'd want it to function on some kind of active feedback mechanism rather than just a timer.
|
|
#
?
Feb 9, 2013 01:47
|
|
|
Parallel Paraplegic posted:Your link loaded the default circuit simulator circuit, but that might just be my browser loving up the link. Anyway I'm wondering, wouldn't you need some kind of interlock? What if the shift doesn't complete in the 75ms, or the mechanism fails - the engine kicks back on at full power and fucks up poo poo? I don't really know much about motorcycles but it seems like you'd want it to function on some kind of active feedback mechanism rather than just a timer. It's not as big of a problem as you might imagine due to how a sequential gearbox works. If you're applying pressure upward on the shift lever, as soon as the transmission is unloaded it will move into the next gear. That said, there are a couple features in some of the commercial products to prevent mis-shifts. They will be disabled below a certain rpm, and they will increase the ignition cut time in lower gears. I'll just have to use the clutch for the first two shifts, and any shifting below 12000 or so rpm. The bike is only used on the racetrack, so that shouldn't be too much of an issue. A mis-shift causes increased wear in the transmission, but a few while dialing the timing in isn't going to ruin anything. 
|
|
#
?
Feb 9, 2013 02:54
|
|
|
edit: solved
wormil fucked around with this message at 23:19 on Feb 9, 2013 |
|
#
?
Feb 9, 2013 04:32
|
|
|
Do you think it's spinning too fast? The treadmil application does require a constant speed so it makes sense to have feedback. Can you find a quick way to rig up the feedback and see if it helps? It looks like a simple RPM sensor. With no feedback it may be driving the motor at full speed and it may not be designed for that. All that said, I'm no motor expert. Magnetics and motors are their own discipline entirely.
|
|
#
?
Feb 9, 2013 19:28
|
|
|
asdf32 posted:Do you think it's spinning too fast? The treadmil application does require a constant speed so it makes sense to have feedback. Can you find a quick way to rig up the feedback and see if it helps? It looks like a simple RPM sensor. With no feedback it may be driving the motor at full speed and it may not be designed for that. All that said, I'm no motor expert. Magnetics and motors are their own discipline entirely. It has speed control. After some thought I suspect the sensor is a speed limiter. But I have no idea what effect it's having by sensing nothing, probably none. Edit: it was an rpm sensor for the heart rate monitor. wormil fucked around with this message at 23:18 on Feb 9, 2013 |
|
#
?
Feb 9, 2013 20:09
|
|
|
Is there a goon consensus on a cheap, not-crap DMM? I think 3 1/2 digit is enough, but a lot of the cheap ones on amazon look suspiciously bad. Would like to spend under $50 or so.
|
|
#
?
Feb 10, 2013 17:42
|
|
|
evensevenone posted:Is there a goon consensus on a cheap, not-crap DMM? I think 3 1/2 digit is enough, but a lot of the cheap ones on amazon look suspiciously bad. Would like to spend under $50 or so. For $60, you can get the not-crap-at-all Extech EX330.
|
|
#
?
Feb 10, 2013 18:22
|
|
|
Zuph posted:For $60, you can get the not-crap-at-all Extech EX330. That's the one I have, it seems to work pretty okay though i don't really have a reference to check it against. That EEVBlog guy reviewed some Extech products (though not this one) and found them to be fairly accurate though "annoyingly and cheaply designed".
|
|
#
?
Feb 10, 2013 18:29
|
|
|
Parallel Paraplegic posted:That's the one I have, it seems to work pretty okay though i don't really have a reference to check it against. That EEVBlog guy reviewed some Extech products (though not this one) and found them to be fairly accurate though "annoyingly and cheaply designed". He actually did review the EX330 as part of his $50 multimeter shootout. https://www.youtube.com/watch?v=ZoeUgMFLyAw It was the overall winner of that video and one of his two recommended meters. The other one he liked was the The Amprobe AM220. The Elenco M-2625 took third and rated a solid acceptable. For what it's worth, I use a EX330 regularly, and would recommend it.
|
|
#
?
Feb 10, 2013 19:33
|
|
|
$47 on Amazon, perfect. Thanks!
|
|
#
?
Feb 10, 2013 20:46
|
|
|
Just an update on that BJT NAND gate I've been trying to build. I finally got a chance to play around with it more, and SnoPuppy's and asdf32's suggestion of putting resistors on the bases of each transistor and getting rid of the unnecessary resistors I had sitting in various places seems to have done the trick. As an added bonus, it didn't even catch on fire and I now know more about transistors and digital logic. Thanks guys!
|
|
#
?
Feb 11, 2013 05:01
|
|
|
So I'm mucking about with a digispark and I've got a program which displays lines of text I send to it over USB. I want to also control an RGB led as a status indicator but I'm running low on output pins. The digispark has 6 pins, two are used for USB data, two are used for the i2c bus, leaving me with one PWM output and one digital pin. I know I can buy a i2c compatible chip that gives me 16 PWM channels, but I really only need to control hue. Is it possible to build a circuit that will take a hue value from the PWM pin and convert it to RGB in hardware? I'd be happy using a 555 timer to generate a clock pulse if needed as it's a lot smaller than a PCA9685 breakout board. I've been googling but haven't found anything other than various PWM emulation libraries and the above breakout board. Edit: Looks like it's possible but it'll be even larger than that board. Could use an analogue to pwm converter circuit (involving a comparator and some basic components I already have per channel, plus a single sawtooth generator) but then I'd need to find some way of converting a pwm signal to three different phases of a non-linear analogue signal. Or could possible do it using a few 555 timers, but again need to split the pwm input into three phases which I haven't worked out how to do yet. Or, something like this: http://www.mhhe.com/engcs/electrical/chapman/fundamentals/PWM.pdf ephphatha fucked around with this message at 03:48 on Feb 12, 2013 |
|
#
?
Feb 11, 2013 06:27
|
|
|
I'm in the market for a cheap LA/oscilloscope. A coworker is pushing this solution, but if another $150 gives me clean & usable software without having to bang around under the open-source hood I'm open to it. I really don't think I need a sampling rate over ~200MHz. Any suggestions?
|
|
#
?
Feb 11, 2013 19:55
|
|
|
JawnV6 posted:I'm in the market for a cheap LA/oscilloscope. A coworker is pushing this solution, but if another $150 gives me clean & usable software without having to bang around under the open-source hood I'm open to it. I really don't think I need a sampling rate over ~200MHz. Any suggestions? I'm saving up for this LA https://www.sparkfun.com/products/10989 More than $200 though, heh.
|
|
#
?
Feb 11, 2013 20:24
|
|
|
JawnV6 posted:I'm in the market for a cheap LA/oscilloscope. A coworker is pushing this solution, but if another $150 gives me clean & usable software without having to bang around under the open-source hood I'm open to it. I really don't think I need a sampling rate over ~200MHz. Any suggestions? I looked through the spec and didn't see anything about the front end wrt. what logic levels it will work at; might want to check if you're looking at anything that isn't common for hobbyists. Incidentally, are you looking for a pure LA (as your coworkers' suggestion seems to be), or something that is useful for analyzing analog signals as well?
|
|
#
?
Feb 11, 2013 21:03
|
|
|
Delta-Wye posted:I'm saving up for this LA https://www.sparkfun.com/products/10989 There is also the cheaper version.. I one at my last two jobs (after convincing them to buy it) and it's great for the price. Bus analyzers with better features exist, but seem to be a LOT more expensive. In fact, we got an expensive one here just for doing I2C/SMBus stuff more effectively. If the software on the Saleae Logic got better, it would be the best thing loving ever.
|
|
#
?
Feb 11, 2013 22:23
|
|
|
Delta-Wye posted:I'm saving up for this LA https://www.sparkfun.com/products/10989 I haven't personally tried one, but a few people in #electronics on Freenode IRC have recommended it. Apparent is capable of working with saleae software and can stream data to PC providing "infinite" buffer. I have the openbench sniffer and the buffer is really freakin pitiful even with RLE enabled, and I don't see any functionality for streaming like I mentioned about the other one. peepsalot fucked around with this message at 22:43 on Feb 11, 2013 |
|
#
?
Feb 11, 2013 22:40
|
|
|
I love this as a cheap $10 multimeter. Had 2 for 2 years. Because it has space for the leads under the cover it's the ideal toolbox multimeter. http://dx.com/p/auto-range-digital-multimeter-9636 Otherwise I recently tried to buy a midrange $40 meter and it failed immediately. Had to fall back on the $10 dx one. Personally, it's going to be $10 or $100 (fluke) from now on. Parallel Paraplegic posted:Just an update on that BJT NAND gate I've been trying to build. I finally got a chance to play around with it more, and SnoPuppy's and asdf32's suggestion of putting resistors on the bases of each transistor and getting rid of the unnecessary resistors I had sitting in various places seems to have done the trick. As an added bonus, it didn't even catch on fire and I now know more about transistors and digital logic. Thanks guys! Great. Slanderer posted:There is also the cheaper version.. I one at my last two jobs (after convincing them to buy it) and it's great for the price. Bus analyzers with better features exist, but seem to be a LOT more expensive. In fact, we got an expensive one here just for doing I2C/SMBus stuff more effectively. If the software on the Saleae Logic got better, it would be the best thing loving ever. Hmm yeah I've seen that one but happened to have gone in this direction: http://www.usbee.com/sx.html Been very happy with it.
|
|
#
?
Feb 12, 2013 01:12
|
|
|
So, in my free time I like to "design" guitar effect pedals/noisemakers. These are all pretty much transistor based. So what I really want to know is what I should be looking for in transistors to determine if they would have any real use.
|
|
#
?
Feb 12, 2013 01:17
|
|
|
My current method has been pretty much look at what pedal parts plus or small bear are selling and ordering that stuff from digikey or whatever.
|
|
#
?
Feb 12, 2013 01:19
|
|
|
I broke a potentiometer connected to a PWM and 120V, 6.2A DC motor (edit: it's a 5K pot). Can I use any basic pot like this? http://www.radioshack.com/product/index.jsp?productId=2102789 2nd Edit: apparently you can and it works as well or maybe better than the original wormil fucked around with this message at 04:53 on Feb 12, 2013 |
|
#
?
Feb 12, 2013 02:13
|
|
|
Stravinsky posted:So, in my free time I like to "design" guitar effect pedals/noisemakers. These are all pretty much transistor based. Depends on what you want to do with it. If it's for distortion, basically you'll just be loving around with different biasing conditions to get the desired results. BJTs are better for hard distortion, JFETs typically have softer distortion over a wide operating range (but if you want no distortion at all you should stick with a BJT). In general though, you can get a huge variety of effects out of any given device, far more difference then you would get between different devices. So don't think too hard about the transistor itself.
|
|
#
?
Feb 12, 2013 03:22
|
|
|
Stravinsky posted:So, in my free time I like to "design" guitar effect pedals/noisemakers. These are all pretty much transistor based. Have a look around http://tagboardeffects.blogspot.com if you haven't found it already. Hundreds of effects clones to choose from, and frequently lots of great discussion in the comments about the merits and pitfalls of various transistors in the circuits.
|
|
#
?
Feb 12, 2013 18:21
|
|
|
So I guess that's a no on sources for reasonably-priced clamp-on inductor probes, then. Should I be able to just take a wide enough ferrite bead, do a couple turns of solid-core insulated wire around it, and then pass the spark plug wire through the eye for a functional induction probe?
|
|
#
?
Feb 12, 2013 19:12
|
|
|
Acid Reflux posted:Have a look around http://tagboardeffects.blogspot.com if you haven't found it already. Hundreds of effects clones to choose from, and frequently lots of great discussion in the comments about the merits and pitfalls of various transistors in the circuits. Yep, its a great site. I have that site near the top of my bookmarks right next to beavis audio research. Though discussion is mainly about what transistor would be a good replacement for whatever clone your building. And I have a severe dislike for building clones. What I really need to do is just stop being lazy and just actually pour over some datasheets. edit: Welp. It looks like Beavis audio research actually put out a cool guide to some transistors. Teaches me to not check back on some of my links for months on end. Stravinsky fucked around with this message at 20:28 on Feb 12, 2013 |
|
#
?
Feb 12, 2013 20:22
|
|
|
|
| # ? May 31, 2024 10:31 |
|
|
http://www.beavisaudio.com/techpages/Transistor-Pinouts/index.htm That is a cool reference. I was collecting some of this information to use as labels in my binder for parts. The problem with "how do I select parts for guitar pedals" is that small details in the transistors can color the sound significantly, complicating choosing one compared to most applications. Hell, a stack of the same make and model can produce different sounds across units.
|
|
#
?
Feb 12, 2013 21:06
|
|