|
Krenzo posted:What's the best way to get a negative voltage supply? I needed +5 and -5 voltage sources for a comparator I'm working with. So, I used a +10V source as the +5, a +5V for ground, and the actual ground for -5V. Is this ok? It works fine now, but I'm about to put multiple devices on a single board that I would like to all use the same ground. Things seem to act up when I start connecting other things to the +5 as a ground. Is there a better way I should be getting -5V? You could make an inverting switcher (Cuk topology or other. It's your preference.) to generate a proper -5V. The issue with doing it the way you are is that the ground return current will have to traverse some rather convoluted paths. Remember that current always flows in a loop. This means that any current going out your virtual +5v to your virtual ground actually has to go through the +10V rail, to ground, and back to the +5v supply (your virtual ground). This is not ideal.
|
#
?
Oct 9, 2011 23:11
|
|
|
|
| # ? Jun 8, 2024 21:31 |
|
|
Krenzo posted:What's the best way to get a negative voltage supply? I needed +5 and -5 voltage sources for a comparator I'm working with. So, I used a +10V source as the +5, a +5V for ground, and the actual ground for -5V. Is this ok? It works fine now, but I'm about to put multiple devices on a single board that I would like to all use the same ground. Things seem to act up when I start connecting other things to the +5 as a ground. Is there a better way I should be getting -5V? Flying capacitor voltage converter. http://www.national.com/mpf/LM/LMC7660.html#Overview If you need more power (say a watt or two) TI and Murata have a lot of small DC-DC converter modules that will work for this. Fake edit: I just noticed that TI bought National Semiconductor. 
|
|
#
?
Oct 10, 2011 00:50
|
|
|
If you're just using the supplies for like comparators and the like, you can probably forget the regulator, since it'll be gumming things up, and use a voltage divider to get your virtual ground. The actual values of your rails shouldnt matter, unless you're not using them as rails, but as references. If you are looking for references, use a dedicated reference generator instead. If you really want an actual negative supply, consider just through on an extra battery or something. Otherwise, a flying capacitor inverter works fine for low current operations.
|
|
#
?
Oct 10, 2011 03:39
|
|
|
Just got my Saleae Logic in the mail, I can't believe I even tried doing MCU development without a logic analyzer! That's something to put in the big post for newbies: if it's digital and has a clock, you need a logic analyzer.
|
|
#
?
Oct 10, 2011 20:41
|
|
|
Hate to double post but I need to get dual power rails into a device I'm building from an external box, so I need at least three pins (four is also fine). It should be a reliable connector that won't be mistaken for something else and preferably something that I can source chassis-mount connectors and plugs for on eBay. Any ideas? E: Low voltage and current, 30V 1A would be a max rating E2: Thanks, I decided on a 4-pin XLR connector, that's commonly used for DC power so I'll make it partially compatible with that (1 is ground, 4 is 12V (15V in my case), 3 will be -15V and pin 2 will probably be something like Enable in future builds) longview fucked around with this message at 09:17 on Oct 14, 2011 |
|
#
?
Oct 12, 2011 22:30
|
|
|
I also got a Saleae Logic a few days ago. Much better software than the old parallel port logic analyzer we were using before. The CANbus analyzer easily payed for this thing in one use.longview posted:Hate to double post but I need to get dual power rails into a device I'm building from an external box, so I need at least three pins (four is also fine). It should be a reliable connector that won't be mistaken for something else and preferably something that I can source chassis-mount connectors and plugs for on eBay. How expensive can they be? An M8 or M12 circular connector might be a good starting point. 
|
|
#
?
Oct 12, 2011 23:08
|
|
|
There are a number of 3 conductor barrel connectors. Also, standard 1/4" TRS would work.
|
|
#
?
Oct 13, 2011 00:13
|
|
|
longview posted:Just got my Saleae Logic in the mail, I can't believe I even tried doing MCU development without a logic analyzer! Wow, that looks pretty awesome, actually. I usually use my scope and attempt to manually decode I2C traffic. Do you have the 8 or 16 channel version? How's the software?
|
|
#
?
Oct 14, 2011 15:44
|
|
|
movax posted:Wow, that looks pretty awesome, actually. I usually use my scope and attempt to manually decode I2C traffic. Do you have the 8 or 16 channel version? How's the software? I have the 8-channel version, I don't expect to need anything beyond that for some time, I understand the 16-channel version is more reliable for high sampling frequencies over time, it has more on-board storage to compensate for jitter in the USB interface. Basically the USB bus is pretty unreliable for real-time data transfer so driver problems or other devices on the same controller can take priority over the Logic data, since the Logic has very limited on board storage it won't last long at higher frequencies. You can actually try the software right now, it includes some demo samples that give you a feel for how it works. There's also an API which lets other programs read data from the device in real-time, and there are some additional protocol decoders in testing on the community site (registration required). Over-all the software is very well written, it's fast, looks good and is very intuitive wrt. defining triggers and protocol analyzers. It can overlay the decoded protocol data onto the waveform like in this screenshot I took: http://longview.be/uploader/uploads/20111014033149PMds1820.PNG It can also export data to a number of formats for further processing. Full disclosure: I haven't really gotten to use it yet, I expect to use it more in the next few weeks when we start doing MCU development in school, but I simply haven't run into any significant issues with the software or hardware for my use yet. The only thing lacking is I guess real-time view of the inputs, turning it into a $150 set of LEDs, but that's been promised for a future software release (the API supports it, just the UI doesn't).
|
|
#
?
Oct 14, 2011 16:37
|
|
|
I'm got a small, dumb question. I'm playing around with IR and have a bunch of these little IR detectors (they look like http://mediagate.pbworks.com/w/page/20682929/f/IR_receiver.jpg) but I don't have a part number/datasheet or anything. Can I rely on the 3 legs being the same Signal, Ground, Vcc? Or does that vary? Also, if I do hook it up wrong will it damage the component at all?
|
|
#
?
Oct 14, 2011 17:20
|
|
|
Trapick posted:I'm got a small, dumb question. I'm playing around with IR and have a bunch of these little IR detectors (they look like http://mediagate.pbworks.com/w/page/20682929/f/IR_receiver.jpg) but I don't have a part number/datasheet or anything. A handful of companies make these guys. Personally, I've only used Vishay ones in the past, but that was primarily due to the free samples (since I wanted to try a handful of really similar ones out before I ordered enough to build 20 units). http://www.vishay.com/ir-receiver-modules/ That list of options is pretty big, but if you look at the datasheets, you'll see that there are actually a dozen or so different ones for each place on that chart. But if you look at the datasheets, you'll find some that, besides the carrier frequency they are designed for, there is a bunch of flexibility between the different models. Basically, you could have anything. Hopefully, it's made for a 36-40kHz carrier, with some amount of gain control and noise rejection, and can work with like RC-5 or something else. But then again, RC-5 has been "interpretted" so many ways, who the gently caress knows. The pinouts will vary, too. The ones whose datasheets i looked at real quick all had the output located on the same pin, but offered options with power and ground switched. Heck, there are probably lots of different options for supply voltage that i didnt notice. Plugging it in backwards probably won't hurt it, unless you overvoltage it at the same time. One of vishay's datasheets has a block diagram that shows a 30kOhm pullup on the output, and an npn transistor with the collector connected to the output, and the emitter connected to ground. So, if you *really* wanted to identify the pins, you could try looking for a pullup in the 1-50 kOhm range on the output. However, if you really want to get something to work, just order a sample from Vishay or single unit from digikey of a component that you know will work with whatever you want to signal it with (even if it's your own microcontroller), and which matches the environment you want it in (in terms of ambient IR noise). That's only really an issue if you want it to work well outside in the sun, though (since you can saturate the internal amplifier).
|
|
#
?
Oct 14, 2011 22:00
|
|
|
Hillridge posted:What is the absolute cheapest thing I can use for potting compound? I had to remove the inch thick jello like coating from the ignition computer in my truck to diagnose a problem, and now I need something to cover all the electronics back up and protect them from moisture. 5140 Methoxy Silicone seems ideal, but it's like $20 a tube. Hot glue is by far the cheapest, not suitable for under bonnet stuff though.
|
|
#
?
Oct 15, 2011 12:40
|
|
|
Hey guys, I've got this board that I'm trying to debug (didn't make it myself), and after hours of pulling my hair out, I just now noticed that my MSP430's crystal oscillator has a pair of capacitors labeled "A223M" on the first line and "Z5U50" on the second line. And, like, I'm not crazy, am I? The "A223" label means that this is a 22 nano farad capacitor with 20% tolerance, not a 22 pico farad capacitor like it is supposed to be?
|
|
#
?
Oct 16, 2011 04:47
|
|
|
I work at a call center, so I need to wear a headset all the time. It would be nice to be able to listen to music through my computer while waiting for my next call. I have a Jabra GN9125 headset wirelessly paired to its base, which plugs into a corded phone via a standard handset wire (4P4C cable). Basically, I want to be able to splice the audio from PC and from the phone line so that I hear both on my headset, but the guy I'm talking to just hears my voice. Is there any ready-made solution for this, or would I need to design it from scratch? If the latter, how would I go about it? Corrupt Politician fucked around with this message at 17:33 on Oct 17, 2011 |
|
#
?
Oct 17, 2011 17:20
|
|
|
Puddy Dumplins posted:Hey guys, I've got this board that I'm trying to debug (didn't make it myself), and after hours of pulling my hair out, I just now noticed that my MSP430's crystal oscillator has a pair of capacitors labeled "A223M" on the first line and "Z5U50" on the second line. And, like, I'm not crazy, am I? The "A223" label means that this is a 22 nano farad capacitor with 20% tolerance, not a 22 pico farad capacitor like it is supposed to be? Yeah, that sounds like 22 nano. It's also a Z5U ceramic material, which will vary by ~50% (or more) over temperature and voltage. For analog circuits where you care about specific capacitance (like oscillators or feedback loops), I tend to stick with X7R or C0G.
|
|
#
?
Oct 17, 2011 19:01
|
|
|
Puddy Dumplins posted:Hey guys, I've got this board that I'm trying to debug (didn't make it myself), and after hours of pulling my hair out, I just now noticed that my MSP430's crystal oscillator has a pair of capacitors labeled "A223M" on the first line and "Z5U50" on the second line. And, like, I'm not crazy, am I? The "A223" label means that this is a 22 nano farad capacitor with 20% tolerance, not a 22 pico farad capacitor like it is supposed to be? Check your datasheets, but I think the MSP430s have integrating capacitance for the crystal inputs. EDIT: It's talked about on p38 of the msp430f1611's datasheet fyi.
|
|
#
?
Oct 17, 2011 22:02
|
|
|
Corrupt Politician posted:I work at a call center, so I need to wear a headset all the time. It would be nice to be able to listen to music through my computer while waiting for my next call. I have a Jabra GN9125 headset wirelessly paired to its base, which plugs into a corded phone via a standard handset wire (4P4C cable). I don't know if anything this specific exists, but you might be able to get it to work with this extremely ghetto method: Get one of those 3.5mm audio jack plugs that you can solder your own wires to at your local electronics store/radioshack/internet, and solder your own wires to it. Say if you're just doing the left channel, figure out what that channel is on the plug, then cut open one of those telephone wires and figure out the same (hold the computer audio wires against the exposed copper of the phone wire until sound comes out the left earpiece) Connect them together through 100ohm resistors as shown  Repeat for right channel You come across as being pretty new to this, so I didn't know how detailed to get. I also don't know how good this will sound, but it's like, 75 cents in parts, so it's totally worth a shot.
|
|
#
?
Oct 18, 2011 03:08
|
|
|
I would do it with amps in between to buffer the two inputs so nothing funny happens.
|
|
#
?
Oct 18, 2011 04:39
|
|
|
So would I, but he wouldn't (assuming he is a stark beginner)
|
|
#
?
Oct 18, 2011 04:44
|
|
|
ante posted:extremely ghetto method Seems like that would only work if the outputs from the two sources use similar voltage ranges. I'm having a hard time finding any information on standards for phone audio, and PC headphone jacks seem to vary somewhat, so I really have no idea. Do phone headsets just use typical audio signals, or is there something weird like a DC bias? At the very least, I'd think that I would have to amplify/attenuate the signal from the sound card so its volume is at a level that makes sense. quote:You come across as being pretty new to this, so I didn't know how detailed to get. I also don't know how good this will sound, but it's like, 75 cents in parts, so it's totally worth a shot. I'm an ECE grad student, but my analog background is a bit meager, and I have far too little experience in actually constructing practical circuits (always been more of a "theory" type guy). So get as detailed as you want. Corrupt Politician fucked around with this message at 06:02 on Oct 18, 2011 |
|
#
?
Oct 18, 2011 06:00
|
|
|
Corrupt Politician posted:Seems like that would only work if the outputs from the two sources use similar voltage ranges. I'm having a hard time finding any information on standards for phone audio, and PC headphone jacks seem to vary somewhat, so I really have no idea. Do phone headsets just use typical audio signals, or is there something weird like a DC bias? Oh, my bad. In that case: summing opamps. Buffers at the inputs will make it stable, and you can use a pot with the summer to adjust the balance of computer/telephone. I have no idea of the standards either, so if you can't find them online or borrow a scope, you'll have to just try it. Headphones are simple as gently caress, so I doubt there's anything weird going on. The wireless bridge is probably just a modulator, with the headphones just being a demodulator and amp. edit: 
ante fucked around with this message at 06:17 on Oct 18, 2011 |
|
#
?
Oct 18, 2011 06:08
|
|
|
SnoPuppy posted:Yeah, that sounds like 22 nano. It's also a Z5U ceramic material, which will vary by ~50% (or more) over temperature and voltage. Yeah, they ended up being 22 fuckin' nF once I clipped 'em off and measured them. Ugh, I am never ever ever getting a board populated by this company again. (this is just yet another of a wide list of problems we've had with them) I mean, to top it all off, these bad caps are so drat tiny that I couldn't even effectively express my rage by throwing them against the wall! They just kinda fluttered around! Puddy Dumplins fucked around with this message at 10:55 on Oct 18, 2011 |
|
#
?
Oct 18, 2011 10:51
|
|
|
I am looking at designing an interface between two pieces of equipment I use regularly at work for automated steering control in agricultural equipment. One system uses a pair of hydraulic valves, so the inputs are either two pwm signals or two analog voltages (depending on the manufacturer). The other system uses a dc motor with what I suspect will be an h-bridge driver. I want to take the two pwm signals or the two analog voltages (one for left, other for right) and use them to drive an h-bridge to run the dc motor. I want to use a microcontroller of some sort to convert the signals, because I want to be able to support both control styles (analog/pwm) with the one unit. In a past life/job I used to code for micros in assembler and later a bit of C. But my experience topped out with the 89c51 series from atmel. I haven't really kept up with the industry, but I know it has come a long way from there! I need some advice on a possible microcontroller to use. I don't have much of a budget at this point, until the project is proven, so I need cheap and easy to program. I need something I can code for in C, and with enough i/o (digital/analog/pwm) to be able to run this, and a whole lot of safety and display stuff as well. I also want something that I can scale - so whatever platform i use for prototyping needs to be something that I can make a hundred or so of for cheap.
|
|
#
?
Oct 19, 2011 05:09
|
|
|
I used atmel ATMega8's in school and they were pretty easy to use. We coded in C but you could do assembly to I think. I don't think they're that expensive either, and they have built in D/A and A/D. If you're looking for something basic with not a lot of bells and whistles that might be the way to go, I know stuff like Arduino has all kinds of inputs for usb and stuff you don't really need. I'm not that knowledgeable about micro controllers beyond that particular one so maybe someone who knows more can add to this.
|
|
#
?
Oct 19, 2011 14:12
|
|
|
I think you will find that things have not changed that much since the 8051 days, except now you can do just fine with free tools. Do you want a module (Arduino style), a dev board, or just a uC? The NXP LPC 17xx series has ADC, PWM capture, and PWM motor control outputs. I haven't use them, but the LPC178x series has an LCD controller. Cost is ~ $7/ea. The LPCXpresso stuff might be about what you need. (There are lots and lots of different micros that will work for your app. I just happened to have used NXP stuff recently)
|
|
#
?
Oct 19, 2011 14:28
|
|
|
niff posted:what arduino starter kit is recommended for someone vaguely experienced with programming concepts and circuit building? really interested in getting back into this sort of stuff, after building some guitar pedals a few years ago. I just got into the whole arduino thing, and since no one really answered. You probably need to have an idea of what you want to do first. At the very least, an Arduino UNO should be fine. There are the normal ones and the SMD ones. The SMD ones dont allow you to remove the processor if you want to wire it into your own circuit board. So you probably just want the normal one. Then you will probably want to get a breadboard or a couple. A jumper wire kit. Some Female to Female jumper wires. Some Male to Male jumper wires. Some hookup wire. A soldering iron (I recommend a Weller adjustable WLC100 as a good beginner), St7 conical tip, solder, solder remover, and random components like switches, buttons, resistors, capacitors, leds, motors, etc. A dremel. A panavise Jr. Helping Hands. Explore your local electronics shop or Radio Shack for some quick purchases or 'whoops i forgot to get this' runs. Which happens about 5 times a week for me. That was just a list off the top of my head from deciding I wanted to build a tracked robot that bumps into poo poo like the rest of the internet. I've been using Robotshop and Cutedigi almost exclusively. They've both been rather quick and cheap (Shipping for cutedigi especially has been like $2/3 a shipment). Philthy fucked around with this message at 19:52 on Oct 19, 2011 |
|
#
?
Oct 19, 2011 19:47
|
|
|
I've noticed this thing, and I'm wondering if anyone else has noticed (seeing how a lot of you are professionals or students in the field). The farther along I get, the more complicated and 'correct' my designs end up being, which greatly increases design time, debug time, etc. Not to mention dealing with it at school all day reduces my desire to play at night. As a result, I feel like the amount I build and tinker for fun on my own time is inversely proportional to the amount I know, or the time I've spent in school, I'm not really sure which. Has anyone else noticed this? It's kind of frustrating and I've been working to fix it lately.
|
|
#
?
Oct 19, 2011 21:23
|
|
|
Delta-Wye posted:I've noticed this thing, and I'm wondering if anyone else has noticed (seeing how a lot of you are professionals or students in the field). I'm a second year electronics student with a significant amount of prior experience, I've spent much more time developing my own designs and prototyping circuits since after I started college. I guess this depends on what kind of work you prefer doing, I usually hack things together and then clean it up later so it's still fun, my design philosophy for personal projects is "that'll do." It probably doesn't help that the electronics program here is set up for people completely new to electronics, this is months into the second year and they've *just* introduced the transistor spending the time until now on diodes (thank god I don't have that class due to prior qualifications, would go mad) and digital counters. I've done no school related independent designing of circuits or programming because none of the classes require it yet.
|
|
#
?
Oct 19, 2011 21:41
|
|
|
I've just finished building a Drone Lab from a kit from Casper Electronics. Here's a video of me messing with it: https://www.youtube.com/watch?v=IY7AWkmYivE I'm still pretty ignorant of electronics theory, hopefully I can pick some up and move beyond building other people's kits.
|
|
#
?
Oct 19, 2011 22:20
|
|
|
Delta-Wye posted:I've noticed this thing, and I'm wondering if anyone else has noticed (seeing how a lot of you are professionals or students in the field). This is from a programming/compsci perspective, I haven't been serious about electronics long enough to burn out yet. I was huge into programming in my spare time from late elementary school through high school, but I found that my desire to do personal projects really dropped off when I was in college. First, I started having way more fun partying. But the big hit was when I got really burned out around my junior year and didn't want to do computer anything. Even though I was considering dropping out of school, I didn't really even notice that I had stopped programming as a hobby. It took awhile after college for me to get back into doing personal programming stuff. I think part of it is natural, you can only spend so much of your time and brainpower on one subject without burning out. And I had been happy trying to go at 110% for years. Also, there is something to what you are saying about making designs 'correct'. When I was 12, I was happy hacking away at code like crazy until it worked. Now I try to "architect" my stuff, even if it is supposed to be for fun. It is hard to have fun when it takes so long to get to something that works, even if everything under the hood is beautiful.
|
|
#
?
Oct 19, 2011 22:35
|
|
|
taqueso posted:I think you will find that things have not changed that much since the 8051 days, except now you can do just fine with free tools. Do you want a module (Arduino style), a dev board, or just a uC? Thanks for the info, yeah the motor controller development kit from NXP looks like it'll do the job nicely. Now I just have to convince the boss to put some $$ up for the project. http://www.embeddeddeveloper.com/tools/2039/Embedded-Artists-AB/EA-XPR-120.htm
|
|
#
?
Oct 20, 2011 01:36
|
|
|
Delta-Wye posted:I've noticed this thing, and I'm wondering if anyone else has noticed (seeing how a lot of you are professionals or students in the field). I have 0 energy to work on personal projects now when I come home since my day job is spent creating schematics, doing PCB artwork, debugging hardware, writing device drivers/FPGA development/etc. I literally turned my hobby into my job and I don't even want to gaze at my workbench when I come home. Though, when I do summon up the energy to work on something, I also over-engineer the poo poo out of it now to the point where it'll take weeks for a project instead of hacking it up in a night. I write up specs, create automated toolchains, comment the poo poo out of the code and everything. 
|
|
#
?
Oct 20, 2011 05:29
|
|
|
Delta-Wye posted:I've noticed this thing, and I'm wondering if anyone else has noticed (seeing how a lot of you are professionals or students in the field). A lot of it has to do with the fact that no matter what you're trying to do, invariably someone makes an IC that already does it, or the design you come up with was thought of by somebody else in the 70s. There's very, very rarely any point to actually building a circuit because if I do a 5 second google search of what I want to do I can find an IC that does it far, far more efficiently than I could have ever done it. Its FUN to design your own power supply, but you can buy a 98% efficient switching power supply off most parts websites for like $30. When I was first learnign it was fun to make diodes light up but there's no reason to make diodes light up anymore, and if there is, like I said there's an integrated package that does it for you already. And with micro controllers being widely available and inexpensive its almost dumb not to use them instead of designing an analog circuit if there's any level of complexity to it.
|
|
#
?
Oct 20, 2011 15:00
|
|
|
Cosmik Debris posted:Its FUN to design your own power supply, but you can buy a 98% efficient switching power supply off most parts websites for like $30. Forgive me if this was just part of a rant, but is this true? 98% seems like it is at the limit of what is possible, much less as a $30 COTS part. My switching design experience has all been for non-isolated wide input range DC-DC buck converters (for example 8-40VDC to 5VDC), so maybe 98% is easy for other types. I've been happy when I can achieve 80-90% efficient over the entire input range. My experience has been that DC-DC converters are actually a good place to do some design work. For example, I have replaced a $15 module with $7 in parts. The new converter is more efficient, cheaper, and has a wider input range.
|
|
#
?
Oct 20, 2011 18:19
|
|
|
Forgive me that was part of my rant, 98% was just a random high number, and yes it depends highly on your specific application. If you are replacing something, it might be better to design your own especially if its a wonky voltage or high amperage you need (I frequently work on vintage keyboards with odd voltage requirements, such as 35V). However in regards to what you said (and correct me if I'm wrong, but what you're talking about is more probably referred to as a switching regulator?), with an 8-40 V input range you are likely better off making your own since the parts would be more on the expensive side and they have to add in the profit margin on top of that. But for real low range applications like 3-7 VDC to 5VDC you can get around a 1 amp IC that's up to 96% efficient for around $3. And I was referring to a AC-DC power supply I bought once which was switching 120AC-24VDC with 90% efficiency for I think $26.
|
|
#
?
Oct 20, 2011 20:10
|
|
|
Ya, I figured that was the case, but was secretly hoping you would tell me about some awesome converters. e: are you saying regulator = DC-DC and converter = AC-DC? My terminology is probably sloppy taqueso fucked around with this message at 20:34 on Oct 20, 2011 |
|
#
?
Oct 20, 2011 20:22
|
|
|
I used to love making web sites for myself until I got a job having to do it for other people. Designing web sites for people who think they're going to get rich selling cheese baskets for Christmas really sucked. Keep your hobbies away from work, etc.
|
|
#
?
Oct 20, 2011 22:04
|
|
|
taqueso posted:Ya, I figured that was the case, but was secretly hoping you would tell me about some awesome converters. This is where it gets tricky. A buck boost converter produces an output that is of higher or lower magnitude than the input (24 volt input to 12 volt output). This can also be referred to as a switching power supply. A power supply is generally what you call the entire stage of a device that, surprise surprise, supplies the power. It might not be anything more than a regulator and a battery. But there might be regulators elsewhere, that are not part of the power supply. BUT a switching power supply is actually a type of switching regulator. Because the device that constitutes a switching power supply is a switching regulator...and a power source. Basically they're all very similar devices. The theory behind all of them is certainly the same. e: oh and yes converters always go from AC-DC and inverters go from DC-AC. Regulators are typically DC-DC and transformers are AC-AC. It is possible to use a switched regulator to go from AC-DC and it is possible to use a regulator to go from DC-AC but I don't think its usually done this way in most applications since the designs are kind of advanced. There's 20 different ways to do anything in circuits. Cosmik Debris fucked around with this message at 02:48 on Oct 21, 2011 |
|
#
?
Oct 21, 2011 02:40
|
|
|
taqueso posted:Forgive me if this was just part of a rant, but is this true? 98% seems like it is at the limit of what is possible, much less as a $30 COTS part. My switching design experience has all been for non-isolated wide input range DC-DC buck converters (for example 8-40VDC to 5VDC), so maybe 98% is easy for other types. I've been happy when I can achieve 80-90% efficient over the entire input range. I found myself needing switching regulators so often, and got tired of paying for the monolithic modules so I just whipped up some little PCBs with my regulator + passives of choice. Switchers are tricky to breadboard, which is unfortunate; that's why the monolithic modules have a market. They're also good for lessening risks on design, instead of screwing up your snubber or something.
|
|
#
?
Oct 21, 2011 05:11
|
|
|
|
| # ? Jun 8, 2024 21:31 |
|
|
Delta-Wye posted:I've noticed this thing, and I'm wondering if anyone else has noticed (seeing how a lot of you are professionals or students in the field). I've noticed something similar. Nowadays the personal projects I get time to work on are just revisions of older designs. So that thing I built last year worked pretty well, but this time I want to give it extra fault protection, a good form-fitting enclosure, feedthrough filter connectors, etc... the core design is mostly unchanged but I'm just paying far more attention to details, and spending tons of time doing so. It's loving tedious as hell and takes a lot of the fun out of it. But my standards have risen so much that I won't feel satisfied with anything else. On the other hand, the stuff I build for my proper research job is the opposite. Everything is a cluster of at least 6 chemical-etched PCBs connected together with a tangle of coax and twisted pair, scattered on a bench. Not to say they aren't well built and functional, but there's no sense nitpicking aesthetics during the development phase (though that might change once I have to start showing the hardware to potential grant donors, which is what my advisor wants).
|
|
#
?
Oct 21, 2011 13:46
|
|
