|
Plus the datasheet says the power only derates to 60%, or exactly 3 watts, if you use none of the heat sink area (ie leave it hanging). I have those resistors too and the literature ratings are usually pretty conservative.
|
#
?
Jul 12, 2010 19:38
|
|
|
|
| # ? Jun 3, 2024 15:07 |
|
|
Zo posted:Plus the datasheet says the power only derates to 60%, or exactly 3 watts, if you use none of the heat sink area (ie leave it hanging). If I'm heating it close to its limit, it'll probably be over 200°C, which is really drat hot. Would I need to worry about the fact that it's soldered onto a pretty tightly-packed breadboard, right next to a bunch of diodes and electrolytic caps? Would it melt the plastic of the breadboard? Am I just being paranoid that such high temperatures make me nervous?
|
|
#
?
Jul 12, 2010 19:44
|
|
|
Corrupt Politician posted:If I'm heating it close to its limit, it'll probably be over 200°C, which is really drat hot. Would I need to worry about the fact that it's soldered onto a pretty tightly-packed breadboard, right next to a bunch of diodes and electrolytic caps? Would it melt the plastic of the breadboard? Am I just being paranoid that such high temperatures make me nervous? Well, I really doubt it'll reach anywhere close to 200C unless you're insulating the heat somehow. It's only 3 watts. I'd rather bolt the resistors somewhere instead of putting on a breadboard though - they are not meant to go on breadboards. edit: is this for the battery powered ultra low duty cycle thing because if so all this is moot Zo fucked around with this message at 20:00 on Jul 12, 2010 |
|
#
?
Jul 12, 2010 19:56
|
|
|
Zo posted:Well, I really doubt it'll reach anywhere close to 200C unless you're insulating the heat somehow. It's only 3 watts. It's for the same project (ultra high duty cycle, actually), but right now I'm just trying to make it work with wall power. When I actually try to make it small and battery powered, I'm going to have to deal with a whole bunch of other heat issues, and I'll design it to be much more efficient. But yeah, where would you suggest I bolt the resistors then? On a separate piece of metal?
|
|
#
?
Jul 12, 2010 20:15
|
|
|
Corrupt Politician posted:It's for the same project (ultra high duty cycle, actually), but right now I'm just trying to make it work with wall power. When I actually try to make it small and battery powered, I'm going to have to deal with a whole bunch of other heat issues, and I'll design it to be much more efficient. I just woodscrew them onto a small plank of wood and keep all my HV stuff together that way. Just prototyping of course. Easy to do and easy to rearrange poo poo if you need to compared to metal.
|
|
#
?
Jul 12, 2010 20:28
|
|
|
Corrupt Politician posted:But yeah, where would you suggest I bolt the resistors then? On a separate piece of metal?
|
|
#
?
Jul 12, 2010 20:30
|
|
|
K then, thanks a bunch. If all goes well, I'll be cauterizing raw meat with a high-voltage square wave before you know it!
|
|
#
?
Jul 12, 2010 20:41
|
|
|
Corrupt Politician posted:K then, thanks a bunch. If all goes well, I'll be cauterizing raw meat with a high-voltage square wave before you know it! Is there much benefit to using electricity over a laser in a fiber optic? It seems like it would be much simpler to use a high powered laser diode that you modulate to control intensity, if all you're trying to do is apply controlled bursts of heat.
|
|
#
?
Jul 12, 2010 20:53
|
|
|
SnoPuppy posted:Is there much benefit to using electricity over a laser in a fiber optic? It seems like it would be much simpler to use a high powered laser diode that you modulate to control intensity, if all you're trying to do is apply controlled bursts of heat. Laser cutting/cauterizing systems certainly exist, but they have some disadvantages, the biggest being that they are much more expensive than the equivalent electrosurgery setup (like $2000 versus $40,000). They also take a lot more skill to use without raising the temperature of the tissue too high and filling the room with smoke and the smell of burning flesh (at least according to a surgeon I talked to). As a result they set the power lower and it takes longer to heat up the tissue. I think it's also partly that doctors are mostly set in their ways and don't like to change tech, and lasers are pretty new. Electrosurgery has been around for like 60 years and is basically the standard; most people don't realize it's used in something like 85% of surgeries.
|
|
#
?
Jul 13, 2010 16:50
|
|
|
Thanks to this thread I've decided to finally bite the bullet and get some proper tools. I've been using a $7 iron for years on cable repairs but I don't think it'll do a satisfactory job with some PCB kits I'm looking at. Any recommendations? I'm hoping to not go over $60 for a decent setup. Currently I'm looking at the Aoyue 937+ Digital Soldering Station ($53 on Amazon) but I'd like some input.
|
|
#
?
Jul 13, 2010 19:56
|
|
|
robotindisguise posted:Thanks to this thread I've decided to finally bite the bullet and get some proper tools. I've been using a $7 iron for years on cable repairs but I don't think it'll do a satisfactory job with some PCB kits I'm looking at. Any recommendations? I'm hoping to not go over $60 for a decent setup. Currently I'm looking at the Aoyue 937+ Digital Soldering Station ($53 on Amazon) but I'd like some input. That looks like a great deal, but I don't have any experience with the construction or long-term reliability of that brand. I don't know how important the digital temperature control would be unless you're working with SMT components - I have never needed it. I use a Weller WLC-100 ($40-50), which is Weller's very basic analog soldering station, and it has been great for everything I've needed. It's definitely a great choice for beginner stuff like through-hole PCB kits and has not given me any problems over years of extended use. Also easy to find replacement tips for.
|
|
#
?
Jul 13, 2010 21:07
|
|
|
robotindisguise posted:Thanks to this thread I've decided to finally bite the bullet and get some proper tools. I've been using a $7 iron for years on cable repairs but I don't think it'll do a satisfactory job with some PCB kits I'm looking at. Any recommendations? I'm hoping to not go over $60 for a decent setup. Currently I'm looking at the Aoyue 937+ Digital Soldering Station ($53 on Amazon) but I'd like some input. I replaced my lovely soldering iron I got at The Source with this unit. For the price I think it is great and you can use real Hakko tips. It has given me no issues so far and being able to turn it on, go wet the sponge and come back to a 300 degree iron in 30 seconds is such a change from waiting 5 mins.
|
|
#
?
Jul 14, 2010 00:26
|
|
|
I have a problem I cannot figure out and I am hoping some goons with actual electronics knowledge can help me out here. My friend gave me some destroyed headphones and I'm trying to connect a new wire to them, but first I need to figure out what wires correspond to what plugs. However it looks like the wires inside the headphone are color coded differently than the part of the cable he still had. Here's what I have so far after several hours worth of work: This chart shows what connectors from the minijack plugs go to what. The center channels share a wire because it's split to go to each side of the headphones. The sub however is not. The stuff on the bottom is just miscellaneous notes.  Next is my horribly lovely diagram of what wires in the headphones go where. Let me explain it a bit. There are 4 speakers in each side of the headphones. 3 actual speakers and one "sub" speaker which I indicate with the three pronged circle. There are also 8 wires coming in from the cable, 7 for audio, 1 for ground.  Click here for the full 2048x1573 image. The diagram on the left has better labels but the diagram on the right has all the solder points the same which is important since many of the wires are soldered on together. Thanks to anyone that can help. If no one can, I have plans to go over to a friends who has an identical working pair and just tracing the signal from the minijacks to the speakers and doing it that way. SpartanIvy fucked around with this message at 08:40 on Jul 14, 2010 |
|
#
?
Jul 14, 2010 08:37
|
|
|
Can anyone give me advice on connecting a surface mount chip without a PCB? I have this somewhat obscure op amp that only comes in a 20-PSOP package, and I want to connect it to a power supply I'm building, but all the surface mount soldering tutorials I find online involve soldering to a custom PCB with a designated slot for the chip. I just want to be able to connect this thing to a breadboard. Here's a picture of the chip:  I'll take whatever option is relatively cheap and easy. Also, I've only done surface mount stuff once before which didn't go too well (and I don't have a particularly steady hand), but I do have access to a microscope and a bunch of good equipment in the ECE shop at my school. Ideally I'd like to buy the DIP version instead (the PA78EU from Cirrus Logic), but I can't find it in stock anywhere online. Corrupt Politician fucked around with this message at 18:26 on Jul 16, 2010 |
|
#
?
Jul 16, 2010 18:21
|
|
|
Corrupt Politician posted:Can anyone give me advice on connecting a surface mount chip without a PCB? I have this somewhat obscure op amp that only comes in a 20-PSOP package, and I want to connect it to a power supply I'm building, but all the surface mount soldering tutorials I find online involve soldering to a custom PCB with a designated slot for the chip. I just want to be able to connect this thing to a breadboard. Here's a picture of the chip: Otherwise I highly recommend one of those custom PCBs, they're like 3 or 4 bucks and your chip is, what, 40-70 bucks? Zo fucked around with this message at 18:50 on Jul 16, 2010 |
|
#
?
Jul 16, 2010 18:47
|
|
|
Sparkfun sells a few surface-mount adapter boards, you might be able to find something that fits the particular chip you're using. Deadbugging it might also be an option.
|
|
#
?
Jul 16, 2010 19:47
|
|
|
Delta-Wye, I think I found a board on proto-advantage: http://www.proto-advantage.com/store/product_info.php?products_id=3100054 It says it's for PSOP-20 chips, but are the dimensions for that package standardized? The spec sheet for the op amp doesn't have any physical dimensions, which is unusual, so I'm left guessing. I don't really want to buy it and find out the pin spacing is slightly off or something. Zo posted:Try emailing digikey to see if they'll sell just one unit of the DIP version (tell them as a sample). Yeah, the chip is pricey, like $42 (the only one I could find with high enough voltage and slew rate that was under $100). I'll try emailing Digikey, I've heard they have really good customer service, probably because they're based in rural Minnesota where everyone is ridiculously polite. Though now that I look at it, the unit price for the DIP version is $31 more, so even if they sold me just one, I doubt it would be worth it (and it would probably be at a higher price because of the low volume). I've never done a custom PCB, they're that cheap? I always thought they were more like $50-100 unless you were buying like a thousand of them. My school has Ultiboard as a PCB layout software, which I've been meaning to learn one of these days. Is there a particular software you'd recommend? What about manufacturers (I know there are a bunch of them online)?
|
|
#
?
Jul 16, 2010 20:43
|
|
|
Corrupt Politician posted:Delta-Wye, I think I found a board on proto-advantage: Sorry, by "custom PCB" I meant those adapter boards. For us, printing our own panel of single layer/dual side PCB will run about $500-600 flat rate, and how many you get depends on the size of your board. But this is for my job so manufacturer and stuff I let our purchaser handle it  The one you linked looks to be standard, 0.05"/1.27mm pitch for PSOP. Should almost definitely work for you chip. I've actually bought stuff off proto-advantage before so I can vouch for them.
|
|
#
?
Jul 16, 2010 21:15
|
|
|
Zo posted:Sorry, by "custom PCB" I meant those adapter boards. For us, printing our own panel of single layer/dual side PCB will run about $500-600 flat rate, and how many you get depends on the size of your board. But this is for my job so manufacturer and stuff I let our purchaser handle it Thanks, I'll get that, then. Now my only fear is ruining a $40 chip in the process...
|
|
#
?
Jul 16, 2010 21:20
|
|
|
Corrupt Politician posted:Thanks, I'll get that, then. Now my only fear is ruining a $40 chip in the process... You could always get someone else to do it for you, depending on how comfortable you are. Also, it's funny to think that for the same price you could get a 2.7Ghz single core CPU. Oh, economy of scale.
|
|
#
?
Jul 16, 2010 21:27
|
|
|
There are quite a few ways to get custom pcbs on the cheap, depending on how quick you need them and what your specific requirements are. http://www.dorkbotpdx.org/wiki/pcb_order He batch processes boards and then orders them in large bunches every couple weeks There is always Sparkfun's https://www.batchpcb.com too, although I have been fairly dissatisfied with their turnaround times. It's not like they don't warn you, but I've experienced 10-week waits. EDIT: I haven't actually used dorkbotpdx, but I got the link and it looks useful. Does anyone have any experience with them? Delta-Wye fucked around with this message at 23:22 on Jul 16, 2010 |
|
#
?
Jul 16, 2010 21:32
|
|
|
Also etching your own pcb is actually really easy, and a simple one-sided board is good enough for the vast majority of hobbyist projects. A lot of people in my class did it for our senior projects in engineering. Lots of good guides online and most of the stuff you need is common household stuff like an iron and laser printer.
|
|
#
?
Jul 16, 2010 21:40
|
|
|
Zo posted:You could always get someone else to do it for you, depending on how comfortable you are. If there were a mass-produced chip that could produce a 200KHz, 300V sine wave, I'd be all over it. It sucks having unusual design requirements.
|
|
#
?
Jul 16, 2010 21:51
|
|
|
Alright, so, if any of you remember my previous post, there's been a further development. In addition to my shitpile of TTL logic ICs, I now have a humorous quantity of surplus vending machine motors. The smallest ones are about the size of an orange, and I've got one that's easily bigger than my head. They've all got very nice all-metal gearboxes. These things all put out a frightening amount of torque, and some of them are pretty high-speed too. So, my question is, what are some wacky projects involving a shitpile of TTL logic and hilariously powerful motors?
|
|
#
?
Jul 16, 2010 23:42
|
|
|
Randel Candygram posted:Alright, so, if any of you remember my previous post, there's been a further development. In addition to my shitpile of TTL logic ICs, I now have a humorous quantity of surplus vending machine motors. The smallest ones are about the size of an orange, and I've got one that's easily bigger than my head. They've all got very nice all-metal gearboxes. These things all put out a frightening amount of torque, and some of them are pretty high-speed too. Robot death machine
|
|
#
?
Jul 16, 2010 23:44
|
|
|
I got my TI 430 Value starter kit today! It's a really powerful little device for the $4 I spent. Now, what to automate? Every coffee machine in the house? RFID Cat Door? Auto-tracking nerf guns?
|
|
#
?
Jul 16, 2010 23:51
|
|
|
It's an MSP430. If you aren't powering it with rotting fruit, you simply aren't doing it right. http://www.youtube.com/watch?v=ZxGZIiyyxrM
|
|
#
?
Jul 17, 2010 00:11
|
|
|
Randel Candygram posted:Alright, so, if any of you remember my previous post, there's been a further development. In addition to my shitpile of TTL logic ICs, I now have a humorous quantity of surplus vending machine motors. The smallest ones are about the size of an orange, and I've got one that's easily bigger than my head. They've all got very nice all-metal gearboxes. These things all put out a frightening amount of torque, and some of them are pretty high-speed too. Robot death machine
|
|
#
?
Jul 17, 2010 00:15
|
|
|
babyeatingpsychopath posted:I got my TI 430 Value starter kit today! I got mine working yesterday, after 2 painful days of stumbling around TI's IAR and CCS IDEs. It's my first time programming a MC (well, aside from my Mindstorms kit 6 years ago). I've never been so happy to see a blinking LED. I'm going to suggest an auto-tracking Nerf gun. That's my plan, at least, so I can keep my desk secure at work. Assuming I can figure out how the hell to do it.
|
|
#
?
Jul 17, 2010 03:06
|
|
|
Corrupt Politician posted:Can anyone give me advice on connecting a surface mount chip without a PCB? I have this somewhat obscure op amp that only comes in a 20-PSOP package, and I want to connect it to a power supply I'm building, but all the surface mount soldering tutorials I find online involve soldering to a custom PCB with a designated slot for the chip. I just want to be able to connect this thing to a breadboard. Here's a picture of the chip: Two ways you can go about it. Flip the chip upside down and hot glue it to a piece of copper clad. Solder small wires to the pins. Then take a razor blade and cut some of the copper clad away so you have isolated pads for the wires to connect to. Otherwise if you can be precise plop the chip down, mark where copper should be missing between the pins and cut that away. Heating the area you want to remove works pretty well to delaminate the copper once you make the cuts. Really unless you absolutely have to use something else just buy a stack of FR4 off ebay for less than the price of a couple adapter boards. You can just cut and hack that to make almost anything you need. Throw those tired old bread boards and vector boards out.
|
|
#
?
Jul 17, 2010 10:25
|
|
|
Corrupt Politician posted:Thanks, I'll get that, then. Now my only fear is ruining a $40 chip in the process... You'd better be very careful then. Heatsink the hell out of it, before you even try powering it up. It's rated thermal impedances are garbage, despite being in a large case. And compensate it so that you only have as much bandwidth and slew rate as required. If this thing becomes unstable, it may destroy itself even without a load. Corrupt Politician posted:If there were a mass-produced chip that could produce a 200KHz, 300V sine wave, I'd be all over it. It sucks having unusual design requirements. ANIME AKBAR fucked around with this message at 14:23 on Jul 17, 2010 |
|
#
?
Jul 17, 2010 14:16
|
|
|
ANIME AKBAR posted:You'd better be very careful then. Heatsink the hell out of it, before you even try powering it up. It's rated thermal impedances are garbage, despite being in a large case. And compensate it so that you only have as much bandwidth and slew rate as required. If this thing becomes unstable, it may destroy itself even without a load. This. Also use a conservatively large current limiting resistor at first. e: I just took a look at the SOA characteristics, make sure you know where you are there! If you're driving from a 300 V bus, then no matter what your AC output voltage is, the average drop across the device is 150 V, which gets you a DC driving ability of 100 mA (at 10% duty cycle!). Without knowing the power factor of the load, the only sensible thing to do is rate the continuous (pseudo-continuous, 10% duty cycle again) current as (300 V / Rlimiter). They also don't supply the pulse width over which this 10% rating is applicable. Of course, if the load current requirement ends up being low enough, you will end up being able to have a big enough limiting resistor for continuous short circuit operation, which would be nice. catbread.jpg fucked around with this message at 01:45 on Jul 18, 2010 |
|
#
?
Jul 18, 2010 01:32
|
|
|
Zo posted:Robot death machine ante posted:Robot death machine Robot death machine it is. This should be interesting to design.
|
|
#
?
Jul 18, 2010 07:15
|
|
|
ANIME AKBAR posted:You'd better be very careful then. Heatsink the hell out of it, before you even try powering it up. It's rated thermal impedances are garbage, despite being in a large case. And compensate it so that you only have as much bandwidth and slew rate as required. If this thing becomes unstable, it may destroy itself even without a load. I completely agree with this. Opamps are not efficient. They're usually designed to give low distortion and noise characteristics, and by doing so they sacrifice efficiency. A raw h-bridge will be a much safer and efficient, plus you can make it from components that don't cost $40 and aren't EOL! You could even switch it at a much higher frequency and use an LC filter to get a nicer sine wave output (class D amp). Then you could use a microcontroller to generate your PWM input and have arbitrary output waveforms.
|
|
#
?
Jul 18, 2010 16:31
|
|
|
I was wondering if you guys could point me in the direction I need. I play sousaphone and would like to attach LEDs to go around the circumference of the bell as well as inside it. What would be the best way to go about doing that? I would like to be able to program the flashing pattern. I was looking at Arduino boards, but I don't know enough about electronics to figure out if they would do the job. The Lily pad model looked like it might work, but I don't know if it can run a lot of lights. The smaller the power unit and board, the better, since I will be dancing and jumping around.
|
|
#
?
Jul 19, 2010 22:47
|
|
|
Brekelefuw posted:I was wondering if you guys could point me in the direction I need. For this, a (555 + 4017) + 10 transistors + 10 resistors + 30 LEDs will do what you want. It should be about $6 or so from Mouser/Digikey/Jameco/whatever. It will run for a week or so on a set of D-cell batteries.
|
|
#
?
Jul 19, 2010 22:49
|
|
|
insta posted:For this, a (555 + 4017) + 10 transistors + 10 resistors + 30 LEDs will do what you want. It should be about $6 or so from Mouser/Digikey/Jameco/whatever. It will run for a week or so on a set of D-cell batteries. How many lights can I run max? 30 seems small for my plans. I am thinking 100+ Also, do the LEDs have to be connected to a board, or can I string them so they can be hung or taped or whatnot. Brekelefuw fucked around with this message at 23:07 on Jul 19, 2010 |
|
#
?
Jul 19, 2010 23:03
|
|
|
Brekelefuw posted:How many lights can I run max? 30 seems small for my plans. I am thinking 100+ Max lights is limited by the transistors you choose. You might want to switch to some power FETs like a BUZ101 or something. Are you wanting ultrabrights or just normal LEDs? That determines power requirements. I'll go into more explanation later after I have initial requirements from you.  The LEDs can be strung around as you desire if they're insulated.I didn't catch the programmability part. The 555+4017 is going to let you hardwire the flash pattern. You can, to an extent, change it but not as easily as a microcontroller.
|
|
#
?
Jul 19, 2010 23:38
|
|
|
I hope I'm not the only person who thought it would be pimp to use the sousaphone as the ground reference. Take that Dallas!
|
|
#
?
Jul 19, 2010 23:59
|
|
|
|
| # ? Jun 3, 2024 15:07 |
|
|
insta posted:Max lights is limited by the transistors you choose. You might want to switch to some power FETs like a BUZ101 or something. Are you wanting ultrabrights or just normal LEDs? That determines power requirements. I'll go into more explanation later after I have initial requirements from you. I don't think ultrabrights are necessary, since the concerts take place in the dark, or with smoke machines etc. At long as they are visible it should be fine. For programmability, I would like to have them flash in a pattern or two. The theme for one of the bands is New Orleans, and I wanted blue and white swirls like a hurricane to spin around the bell like a toilet flushes (bad explanation?) Or I would like to put a solid logo up for the other band. Sound reactivity would be good as well once I learn how to do the initial idea. I guess I would need a controller to tell which pattern to go to etc.
|
|
#
?
Jul 20, 2010 03:54
|
|