|
asdf32 posted:I doubt you can guarantee the polarity of the pulse so I'd stay with ceramic. Probably pretty high voltage too, maybe 100+ (though I didn't look at this specific application). On top of that I'd suggest diode protection - you say it's going into a 555 so have a diode going from this pin pointing twards Vcc and from ground pointing to this pin. Schottky. That will clamp transients to the rails and help protect the part. I'd also have a resistor in series with the cap to reduce the current. Also have good bypassing on Vcc to absorb these transients, at least 1uf, probably 4.7uf. Is this not what the 100K and 1N4148 are doing in diagram I linked? It looks to me like it's rigged to allow overvoltage to cause breakdown in the diode and dump out onto the common ground, with the resistor presumably there to calm down any excess amperage. If this is not the case, what is the function of that pair? And to make sure I understand what you're telling me: I should have a diode allowing flow from the In(-) to Vcc, another diode set up to break down in an overvoltage condition and then allow flow from the In(-) to ground, a resistor(just any, so long as the resultant voltage after it meets or slightly exceeds trigger voltage?) in series after the 22pf cap, and lastly a cap between Vcc and the common circuit ground of several microfarads? I get the diode and resistor, and the breakdown diode between In(-) and common ground; I don't quite understand why I would stick a cap between Vcc and ground. Is the cap being used as an energy sponge that then dumps excess current back out to the negative common ground and into the DC source? It looks at first glance to me like it would function as an intermittent short in the circuit, the "energy sponge" thing is the only explanation that springs to mind as to why it is desirable to do this as protection. I tried to look up the voltage going out to the plugs, unfortunately it only gives the peak voltage that should go into the low-voltage side of the coil packs. 100 V is what it's supposed to top out at, taking a wild guess I'd think that 60KV or more is going down the plug wires to fire the spark. However I'll just test a plug and find out for sure, then build for bit higher than tested value. When you're talking about using diode protection, you mean Zeners right? I was under the impression regular diodes fry after a couple voltage breakdown conditions occur. Good to know about the tantalums; I just went and read about them a bit, I wonder if a miniature version of a thermite reaction isn't happening when they go overcurrent and blow. I certainly wouldn't want to
|
#
?
Jan 31, 2013 06:38
|
|
|
|
| # ? May 31, 2024 23:32 |
|
|
Bad Munki posted:To elaborate a bit, the max level of shock I'm going for here would be, say, the feeling of a nice fresh 9-volt battery on the tongue. It doesn't feel good, might even say it hurts if you leave it on there for a bit, but it's also not something I would be at all concerned about causing cardiac arrest ever, or even burning the flesh, as the pain reflex tends to remove the contact long before it becomes problematic. Sure, one could choose to ignore the pain and let a 9-volt do some tongue cooking, but that's sort of irrelevant here. We're assuming a normal response. Solution: Make a small joystick, make them play with their tongues. EDIT: 
|
|
#
?
Jan 31, 2013 07:05
|
|
|
It's not quite as bad as the guy who wanted help with his homemade TENS machine, but come on. If you need to ask for help on how to electrocute people, don't loving do it.
|
|
#
?
Feb 1, 2013 00:39
|
|
|
Kilersquirrel posted:Is this not what the 100K and 1N4148 are doing in diagram I linked? It looks to me like it's rigged to allow overvoltage to cause breakdown in the diode and dump out onto the common ground, with the resistor presumably there to calm down any excess amperage. If this is not the case, what is the function of that pair? And to make sure I understand what you're telling me: I should have a diode allowing flow from the In(-) to Vcc, another diode set up to break down in an overvoltage condition and then allow flow from the In(-) to ground, a resistor(just any, so long as the resultant voltage after it meets or slightly exceeds trigger voltage?) in series after the 22pf cap, and lastly a cap between Vcc and the common circuit ground of several microfarads? So first, I didn't see the diagram you linked. Addressing what I said anyway though, yes I was describing two diodes - one would allow current to flow when voltage at in - exceeded Vcc, and the other would allow current to flow when in- went below gnd (this wouldn't be breaking down - it would still be forward biased with GND on one side and -0.7V or so on the other). This is one type of protection scheme. However the diagram is a good way to go. The BJT is isolating the input from the 555 and providing protection. That diode is doing the second half of what I described - clamping any negative transients to ground. The 100k resistor is helping to hold the cap at GND level and dissipate some of the charge within it. Note that a BJT essentially contains an internal diode that can be seen as being in paralel with the one shown, but in the opposite direction. Between these two diodes (one shown, one in the BJT) the BJT input will always remain clamped within roughly 0.7V of GND and this will be pretty well protected. The BJT turns on when current flows across the internal diode which will happen when the voltage on the spark plug rises quickly. So to sum it up, stick with the diagram. quote:I get the diode and resistor, and the breakdown diode between In(-) and common ground; I don't quite understand why I would stick a cap between Vcc and ground. Is the cap being used as an energy sponge that then dumps excess current back out to the negative common ground and into the DC source? It looks at first glance to me like it would function as an intermittent short in the circuit, the "energy sponge" thing is the only explanation that springs to mind as to why it is desirable to do this as protection. The cap I was describing would be called a bypass cap. You're basically right, it's a local reserviour of charge. It can absorb local transients, or supply current if voltage dips. Basically it just helps regulate voltage right there at the chip's pins where it's needed. It's not strictly necessary, but definitely a good idea. Ideally I'd probably have a 0.1uf and a 1uf or 4.7uf in this noisy application, but consider anything between 0.1uf and 4.7uf. quote:I tried to look up the voltage going out to the plugs, unfortunately it only gives the peak voltage that should go into the low-voltage side of the coil packs. 100 V is what it's supposed to top out at, taking a wild guess I'd think that 60KV or more is going down the plug wires to fire the spark. However I'll just test a plug and find out for sure, then build for bit higher than tested value. When you're talking about using diode protection, you mean Zeners right? I was under the impression regular diodes fry after a couple voltage breakdown conditions occur. If it's 60kv you're screwed, you can't buy 60kV caps realistically, I'm sure it's not. No I wasn't talking about zeners because I wasn't talking about reverse breakdown. See this link for what I was talking about although again, I don't think it's necesary to add this with that BJT circuit. http://www.beam-wiki.org/w/images/7/7d/Fig_5_-_Internal_Protection_diodes.GIF
|
|
#
?
Feb 1, 2013 00:47
|
|
|
I was hoping I could get some quick advice here. So I have a headset (single ear cup and boom mic) that was a 2.5mm jack but quite frayed. I have a spare cord with a 3.5mm jack on it so I thought I'd graft them together but when I stripped the wires I got this:  I'm an idiot, what do I do!? Is there a better thread to ask in? I think there used to be a DIY electronics thread but I can't find it.
|
|
#
?
Feb 1, 2013 02:07
|
|
|
Moist von Lipwig posted:I was hoping I could get some quick advice here.  I would just measure the cut off plugs and match the cable up with the conductor it used to go with. Moist von Lipwig posted:Is there a better thread to ask in? I think there used to be a DIY electronics thread but I can't find it. Congratulations, I think you found it again.
|
|
#
?
Feb 1, 2013 02:34
|
|
|
EDIT: ^^ Basically what they said ^^Moist von Lipwig posted:I was hoping I could get some quick advice here. That copper twisted around the white wire is the ground line which doubles as a shield against interference, some cables have them and some just put the ground line in another wire (probably the white or black wire in the other one, but I'm not certain). If you have access to a continuity checker, you could use that to find out which wires go to which part of the jack (tip, middle or base). Then, assuming you kept the old jack you cut off, do the same to that one to figure out which wires are which and wire up accordingly. If you have a multimeter you could use that as well, or you could probably pick up a cheap continuity checker at Radio Shack, they're just little things that beep or light up when they detect a complete circuit. I wouldn't guess at this, you'd run the chance of shorting out whatever you plug it into (unless someone else here already knows exactly which wires go to what, I have to check whenever I hack a cable apart just for peace of mind  )Also it looks like you're wiring a mono headset with microphone to what is usually assumed to be a stereo headphones-type jack (which is fine, it will work) but make sure whatever you plug it into is not expecting stereo headphones and is in fact designed to have a mono headset with microphone plugged into it - otherwise it will try to run sound signals through the microphone and probably destroy it. If you don't need to use the mic you might want to just leave that completely disconnected.
|
|
#
?
Feb 1, 2013 02:37
|
|
|
asdf32 posted:Schooling me Ah, I understand. I missed that you hadn't looked at the diagram somehow, I thought you were stating additions I should have. Thank you for breaking that all down for me, I understand the reasoning for what you were suggesting for input protection much better now, and the mechanics of it. As for the voltage, I was taking a wild stab at the voltage going down the plug wire rather than what will be coming down from the pickup. If there was enough voltage to send 60KV down the probe I'd be shooting lighting bolts out of my coil packs, I think. It probably is much less than 60KV going through the plug wires, though. I'll find out this weekend. For the pickup, will a solid-core wire wound on the coil work decently as a pickup, or will I definitely need a manufactured pickup? I'm not opposed to either but nobody seems to carry clamp-on pickups on their own, and I'm trying to avoid having to mail-order things unless absolutely necessary. I work right by a Fry's so I should be able to get most anything I need, and make up the rest from Radio Shack. I'll be buying most of the parts sometime this weekend(hooray, payday) and reacquainting myself with how to solder properly by disassembling and re-soldering some old junk electronics.
|
|
#
?
Feb 1, 2013 04:36
|
|
|
Kilersquirrel posted:Ah, I understand. I missed that you hadn't looked at the diagram somehow, I thought you were stating additions I should have. You could add a coil or some ferrite rings around your wire to shave of the worst transients, if you are worried about your caps that much. Kilersquirrel posted:For the pickup, will a solid-core wire wound on the coil work decently as a pickup, or will I definitely need a manufactured pickup? I'm not opposed to either but nobody seems to carry clamp-on pickups on their own, and I'm trying to avoid having to mail-order things unless absolutely necessary. I work right by a Fry's so I should be able to get most anything I need, and make up the rest from Radio Shack.
|
|
#
?
Feb 1, 2013 11:08
|
|
|
Is there any sort of rotary switch where each position will make an additional contact while maintaining all the previous contacts? IE first position is open, second position closes one circuit, third position keeps the previous circuit closed and closes a second circuit, and so on.
|
|
#
?
Feb 1, 2013 22:02
|
|
|
I'm not sure if that exists if what you want is for say position 4 to make contacts 1, 2, 3 and 4, position 2 makes 1 and 2 etc. but depending on the exact application it might be possible to design some logic to implement it with a normal rotaty switch/encoder?
|
|
#
?
Feb 1, 2013 22:16
|
|
|
What if you had a rotary switch with discrete positions and you just put diodes between each position, pointing "downhill?" Wouldn't that effectively do what you're asking?
|
|
#
?
Feb 1, 2013 22:17
|
|
|
TerminalSaint posted:Is there any sort of rotary switch where each position will make an additional contact while maintaining all the previous contacts? IE first position is open, second position closes one circuit, third position keeps the previous circuit closed and closes a second circuit, and so on. What are you trying to do? You can also use a potentiometer and a bunch of comparators as well.
|
|
#
?
Feb 1, 2013 22:21
|
|
|
What type of signal do you want to connect through the switch?
|
|
#
?
Feb 1, 2013 22:21
|
|
|
Specifically I'm making a pretty simple circuit that has several LEDs in parallel. I'm building a prop replica that will feature a knob on the side anyway, and it occurred to me that if such a switch existed I could have the knob control overall brightness (by turning off or on individual LEDs). If they were incandescent bulbs I'd just go with a variable resistor; but as I understand it, LEDs have a much smaller range of brightness, correct? Edit: Space is already pretty tight so if I could get it in a single switch I might be able to cram it in. I wouldn't have room to hack together much of a workaround. TerminalSaint fucked around with this message at 22:40 on Feb 1, 2013 |
|
#
?
Feb 1, 2013 22:38
|
|
|
Bad Munki posted:To elaborate a bit, the max level of shock I'm going for here would be, say, the feeling of a nice fresh 9-volt battery on the tongue. It doesn't feel good, might even say it hurts if you leave it on there for a bit, but it's also not something I would be at all concerned about causing cardiac arrest ever, or even burning the flesh, as the pain reflex tends to remove the contact long before it becomes problematic. Sure, one could choose to ignore the pain and let a 9-volt do some tongue cooking, but that's sort of irrelevant here. We're assuming a normal response. Hook a 9V to an automotive relay (5 pin SPDT, or a 4pin SPST normally closed if you can find one.) You basically hook one battery terminal to one side of the coil, then wire the other side of the coil through the normally closed circuit, and then into the battery. When the coil energizes, it opens the circuit, denergizing the coil causing the switch to close again, and oscillate. Every time the switch is opened the inductive kickback makes enough voltage to give a jolt. And it makes a fun buzzing sound so you know it's live! The coil size of the relay will determine how much shock you get from it. I tried a small pcb mount relay first and could only just barely feel any tingle only on some parts of skin, trying very hard. With the automotive relay it's enough to feel pretty easy, but not overwhelming. peepsalot fucked around with this message at 22:43 on Feb 1, 2013 |
|
#
?
Feb 1, 2013 22:40
|
|
|
TerminalSaint posted:Specifically I'm making a pretty simple circuit that has several LEDs in parallel. I'm building a prop replica that will feature a knob on the side anyway, and it occurred to me that if such a switch existed I could have the knob control overall brightness (by turning off or on individual LEDs). If you want to turn on a certain number of LEDs, you could do it with a single rotary switch. Connect all the LEDs together in a string, connect V+ to one end. At the junction between each set of LEDs and and the end of the string, connect to one of the switch legs through an appropriate resistor. The common leg of the switch connects to ground. When the switch is totally open, there is no path to ground, so no LEDs are on. When the switch is on, one of the resistors will connect to ground. All the LEDs after that point will be off, the LEDs before that will be on with their current set by the resistor. V+ will need to be higher than the sum of Vfwd for the entire string. (This is basically what Bad Munki was talking about, but the diodes are your LEDs.) I don't think LEDs have a lower brightness range if PWM dimming is done well. You could probably get a small flashlight control board from candlepowerforums marketplace that will dim a few LEDs with potentiometer control. taqueso fucked around with this message at 22:50 on Feb 1, 2013 |
|
#
?
Feb 1, 2013 22:48
|
|
|
taqueso posted:If you want to turn on a certain number of LEDs, you could do it with a single rotary switch. Connect all the LEDs together in a string, connect V+ to one end. At the junction between each set of LEDs and and the end of the string, connect to one of the switch legs through an appropriate resistor. The common leg of the switch connects to ground. When the switch is totally open, there is no path to ground, so no LEDs are on. When the switch is on, one of the resistors will connect to ground. All the LEDs after that point will be off, the LEDs before that will be on with their current set by the resistor. V+ will need to be higher than the sum of Vfwd for the entire string. (This is basically what Bad Munki was talking about, but the diodes are your LEDs.) Ah, that makes sense, though I'll have to diagram it to fully grasp it. I think it will fit the bill if I can find a small enough switch. As far as PWM, each of the parallel circuits includes a candle flicker LED with a built in flicker IC, so I'm not sure if that would prove problematic(my grasp of electronics is tenuous at best).
|
|
#
?
Feb 1, 2013 23:02
|
|
|
peepsalot posted:Hook a 9V to an automotive relay (5 pin SPDT, or a 4pin SPST normally closed if you can find one.) You basically hook one battery terminal to one side of the coil, then wire the other side of the coil through the normally closed circuit, and then into the battery. When the coil energizes, it opens the circuit, denergizing the coil causing the switch to close again, and oscillate. Every time the switch is opened the inductive kickback makes enough voltage to give a jolt. And it makes a fun buzzing sound so you know it's live! Using inductive kickback to generate the spark is an interesting idea. I was thinking isolating the shocking circuit from the rest of the electronics would give you a bit of safety factor and should keep any stray current paths from causing surprise damage.  All you need to do is turn the LED on momentarily and the sparkgap (or person) should discharge the inductor. You can even use the optoisolator to trigger a power FET if you can't find a optoisolator that can handle a near short to ground without smoking.
|
|
#
?
Feb 1, 2013 23:45
|
|
|
Delta-Wye posted:Using inductive kickback to generate the spark is an interesting idea. I was thinking isolating the shocking circuit from the rest of the electronics would give you a bit of safety factor and should keep any stray current paths from causing surprise damage. The isolation is definitely a good idea. An opto is only good for 10's of miliamps usually but the bigger problem would be the voltage - it's exposed to the full inductive spike (which would happen when the opto shuts off). I think it needs a pretty high voltage mosfet (500+V) to work. Although that's not too hard to find. TerminalSaint posted:Ah, that makes sense, though I'll have to diagram it to fully grasp it. I think it will fit the bill if I can find a small enough switch. So you're going to have a bunch of flickering LED's on at the same time? That would be kind of a weird pattern of light and yes, it would prevent you from dimming them. If you're talking about LED's directly then there's no problem ranging their brightness - you just need to vary a resistor or get a current controlled led driver (not too hard to find). @peepsalot Nice idea, it sounds like you've done this?
|
|
#
?
Feb 2, 2013 01:32
|
|
|
TerminalSaint posted:Ah, that makes sense, though I'll have to diagram it to fully grasp it. I think it will fit the bill if I can find a small enough switch.
|
|
#
?
Feb 2, 2013 15:19
|
|
|
These ones are pretty interesting, they've actually got the IC built into the LED: Not exactly a preamp, but here's a similar one hooked up to a little magnetic buzzer. https://www.youtube.com/watch?v=753-lkao8l0&t=61s
|
|
#
?
Feb 2, 2013 18:51
|
|
|
What do you guys think of the Maker kits? I'm a total electronics newbie looking to learn and they sell the book and all of the bits you'll need for it for $239. Is this a good value, or would I do better tracking everything down myself? http://www.makershed.com/Make_Electronics_The_Complete_Collection_p/mecp4.htm?CartID=1
|
|
#
?
Feb 2, 2013 22:47
|
|
|
Those component packs look like pretty ripoff prices to me. You could go on mouser and get all that poo poo for probably $50 or something. I mean I guess it's nice that it's all prepackaged and layed out as a kit, but it depends how much you value your time of finding and buying all that stuff yourself. You can get resistor packs with 20x-100x of every common value for $10-$20 on ebay or whatever like this http://www.ebay.com/itm/1-10M-1120p...=item1c29f16a51 or this http://www.ledsee.com/index.php/resistor/e12-range-resistor-set-1-8500-pieces-detail capacitors http://www.ebay.com/itm/1uF-2200uF-...=item19d4ef9d1f ,heatshrink tubing http://www.harborfreight.com/127-piece-heat-shrink-tubing-set-67524.html same with jumper wires, connectors, etc. peepsalot fucked around with this message at 23:34 on Feb 2, 2013 |
|
#
?
Feb 2, 2013 23:27
|
|
|
peepsalot posted:Those component packs look like pretty ripoff prices to me. You could go on mouser and get all that poo poo for probably $50 or something. Yeah, I've found that a lot of people who don't do electronics don't realize that components are mass produced to such levels that a lot of them have absurdly low prices. Some of them are fractions of a cent each. If you go to mouser or jameco etc and just buy the stuff that's in that kit you can probably save a bunch. My last order had about 400 total quantity of items and cost me about $55 shipped. Then just get yourself a good breadboard and some kind of power supply (I use a wall wort that I run through a little board with a 5V regulator and some capacitors on it, really simple to build - here's the same basic circuit though I put bigger caps on it: http://www.instructables.com/id/5v-Regulator/). If I were you I'd just buy the book by itself, then get the parts separately. I definitely recommend buying some of those grab-bags for capacitors and resistors and LED's etc like peepsalot said. Jameco has them, Amazon has them (though it's more expensive from them) or you can get the eBay ones. It's great for fleshing out your supply of some of the most frequently used parts, and they usually come in the most frequently used values. Now keeping all those parts sorted is a bit of a bigger problem. I keep my parts in a big tackle box, but this probably isn't ideal since it's not anti-static. You can buy specifically anti-static containers but they're a bit more. Still probably cheaper than that kit.
|
|
#
?
Feb 2, 2013 23:35
|
|
|
My Dreamcast VGA mod is coming along. I tried skipping the capacitors on the RGB wires (see here) and I don't see any difference in the video quality. There's a little interference though, probably from the way I'm using a ribbon cable: Any ideas on what I should do to try to minimize interference? The DE-15 connector I'm using is on an 8" ribbon cable itself, how big of a difference would shortening that make?
|
|
#
?
Feb 3, 2013 06:50
|
|
|
I spent a few hours trying to part my way though the first kit and it was just way too time consuming - I know next to nothing so the minor variations between different parts were incomprehensible to me. I bet after I get through the book it'll all be really obvious but at this point I just don't have time to guess at it and wait around. I need to learn electronics now, not sometime next month, I'm just going to write it off as a business expense. Any of you guys ever work with laser projectors? I'm modifying a couple of them right now, the parts are all analog and it's pretty modular, like building a PC tower... learning electronics will help fill out my skill set when dealing with them. I can share some stuff as I do it if anybody is interested in the internal mechanics of how they work, I've got a couple small projects I need to do to get them up to code.
|
|
#
?
Feb 3, 2013 14:23
|
|
|
What are you trying to accomplish exactly? Because if you think you will have a good grasp of electronics immediately I have some bad news for you. Unless your some sort of savant, its going to take time. Edit: like if you need to know how to do just a thing just ask and someone will more than likely tell you how if its that important to get it done now. Stravinsky fucked around with this message at 20:08 on Feb 3, 2013 |
|
#
?
Feb 3, 2013 20:06
|
|
|
Well, I redid my Dreamcast VGA circuit on some proper prototyping board and the ghosting went away. Maybe solderless breadboards aren't the best at carrying video signals  Also, I did some research and apparently a lot of RGB cables for game consoles use 220uF caps as a DC offset filter because RGB needs a clean 0V signal for black. I've seen plenty of VGA cards without filter caps on the output so the VGA spec must take care of DC offset some other way. This means I can just skip them in my final design and make an absolutely tiny board that can fit wherever I want the VGA port to go.
|
|
#
?
Feb 3, 2013 21:29
|
|
|
So I'm trying to figure out what the heck I'm doing wrong here. I'm trying to recreate this circuit for a CMOS NAND gate, only using BJT's. So this circuit but with discrete PNP and NPN BJT transistors: Everything seems to be going fine if I build everything (connecting A and B to a pull-down resistor connected to ground) but leave out one of the PNP's. However, the second I connect the missing PNP (which could be either one) the LED I'm using to indicate the output state dims to about half of its normal brightness and the PNP heats up really fast, actually burning my finger and releasing the smoke in one case where I didn't pull it out fast enough. I have no idea how I'm doing this, I have 1k resistors on every possible path to ground (I even replaced them with 100k and it didn't make a difference) and the resistors don't seem to heat up much at all. Some permutation of this circuit should be possible to build with discrete BJT components, right? Am I missing some fundamental "catch on fire" property of BJT's that I don't know about that prevents them from working in this circuit? Shame Boy fucked around with this message at 03:07 on Feb 4, 2013 |
|
#
?
Feb 4, 2013 03:00
|
|
|
Parallel Paraplegic posted:So I'm trying to figure out what the heck I'm doing wrong here. I'm trying to recreate this circuit for a CMOS NAND gate, only using BJT's. So this circuit but with discrete PNP and NPN BJT transistors:http://forums.somethingawful.com/editpost.php?action=editpost&postid=412192111 Hard to say 100% without seeing the schematic as built, so, do you have PNP transistors above, emitters to Vdd, and NPN below, B's emitter to Vss, and the two bases B connected directly together? On a bipolar transistor, the base-emitter junction looks like a diode, so if so, you'd basically have two diodes between Vdd and Vss, even with all the transistors off. With MOSFETs, the gates are insulated and don't source or sink current whether the transistors are on or off (they require charge while switching on or off, like charging a capacitor) which is a pretty significant difference.
|
|
#
?
Feb 4, 2013 03:14
|
|
|
Do you have a diagram of the actual circuit you're building, with BJTs?
|
|
#
?
Feb 4, 2013 03:15
|
|
|
Parallel Paraplegic posted:So I'm trying to figure out what the heck I'm doing wrong here. I'm trying to recreate this circuit for a CMOS NAND gate, only using BJT's. So this circuit but with discrete PNP and NPN BJT transistors:http://forums.somethingawful.com/editpost.php?action=editpost&postid=412192111 First, post your exact schematic. It helps with debug, since guessing how you've implemented the circuit is likely to be wrong. There's a reason why this was done with CMOS, and not BJTs. With a BJT, you will need to have some base resistance or you will effectively have a diode between the base-emitter, and the base-collector. A MOSFET doesn't have this problem because the gate is effectively a capacitor. Once the gate is charged up, no more current can flow. If you really isolated all power/ground paths with 1k resistors, nothing should be able to get hot unless you're using a very high voltage. Even a 1K short between +5 and ground would only be 25 mW of power. I suspect you have some other path to ground, or some other voltage rail, that you're not accounting for.
|
|
#
?
Feb 4, 2013 03:20
|
|
|
Yeah I should have posted a schematic, right now I'm just dicking around on the breadboard but I guess "hey guys my thing that kind of looks like this isn't working, help!" is a pretty dumb thing to post. Sorry!  Anyway I had missed a path to ground that I didn't see until I carefully went back over what I had done. Somehow I missed it in all the previous goings-over I did. I am a dumb again, I guess. It's not heating up anymore but it's still not working, probably due to the several reasons you have already pointed out. I did actually have resistors on the bases. I've put an LED between the NPN's and ground to see if they're actually turning power on and off, and they are, but the voltage across the output never changes. I tried lowering the value of the resistor on the NPN's output thinking that if it had a preferential path to ground it might at least drop the voltage across the output, but still nothing. I will draw up a schematic and get back to you guys, sorry I didn't do that to begin with
Shame Boy fucked around with this message at 03:48 on Feb 4, 2013 |
|
#
?
Feb 4, 2013 03:45
|
|
|
Okay, here's what I drew up, including the values of resistors I currently have in there while loving around with it. If I switch A and B to low, the MONITOR LED is off and the OUT LED is on. If I switch them both to high, MONITOR comes on, but OUT remains on, and the voltage across OUT remains totally unchanged.
|
|
#
?
Feb 4, 2013 04:28
|
|
|
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. I think the problem is the bases of the PNPs and NPNs are connected together. This provides a path for current to flow from the PNP emitter through the base, and into the 220 ohm and base of the NPN transistor, turning it on. It is generally a very bad thing to have high base currents. Try sticking a 1K on the base of each transistor and see what happens. edit: After looking at it again, and reading your post a second time, I think that you have the base-collector junction of your NPN transistors forward biased when the inputs are pulled high. This causes current to flow through the base to the collector and turn on the LED. SnoPuppy fucked around with this message at 04:39 on Feb 4, 2013 |
|
#
?
Feb 4, 2013 04:35
|
|
|
SnoPuppy posted:I think the problem is the bases of the PNPs and NPNs are connected together. This provides a path for current to flow from the PNP emitter through the base, and into the 220 ohm and base of the NPN transistor, turning it on. It is generally a very bad thing to have high base currents. 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?
|
|
#
?
Feb 4, 2013 04:52
|
|
|
Can someone point me to some good sequential tutorials for mastering the MSP430GXXX chips? 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. Also, what's the difference in using a relay vs. a BJT for current-controlled current source? My beaglebone can supposedly only source 4-6mA from each GPIO so I'm wondering if BJTs are the way to go for controlling LEDs and so forth.
|
|
#
?
Feb 4, 2013 05:24
|
|
|
Kire posted:Also, what's the difference in using a relay vs. a BJT for current-controlled current source? My beaglebone can supposedly only source 4-6mA from each GPIO so I'm wondering if BJTs are the way to go for controlling LEDs and so forth. BJTs (or logic level MOSFETs) are absolutely the way to go. Relays that will run at that low of voltages and current are rare. I did a quick search on digikey for a 3.3volt relay that only took max of 6ma to run, and found a grand total of 1 relay that would actuate. It was $3.70 each. In addition, even if you had such a low power relay, the inductive kick that the coil will generate could easily damage the beaglebone. Don't run low power motors directly off I/O pins for the same reason. If you ever have a project that needs a bunch of driver transistors, but not a huge amount of current per channel, I'd like to point you at the ULN2803. It's 8 darlington transistors on a single I/C. It's good for 500ma per channel, but if you're driving them all (and you're using an 8 channel for a reason) you'll need to reduce per channel current. It's a little opaque to figure out by how much though. (power dissipation graphs!) Either way though, it can be mighty convenient to only have to deal with one ic instead of 8 transistors.
|
|
#
?
Feb 4, 2013 06:50
|
|
|
|
| # ? May 31, 2024 23:32 |
|
|
Silver Alicorn posted:Well, I redid my Dreamcast VGA circuit on some proper prototyping board and the ghosting went away. Maybe solderless breadboards aren't the best at carrying video signals Monitors have input coupling caps and DC-restoration is a standard feature. You should remember that ideally each colour video line has its own ground, and proper VGA cables use a separate coaxial cable with separate ground for each R G B signal.
|
|
#
?
Feb 4, 2013 11:07
|
|