|
ate all the Oreos posted:I simulated pulling the Reset pin down in LTSpice and it didn't turn the oscillator off, it just made it so that the second it went high it went low again (so like a 1% duty cycle). Admittedly this is probably just LTSpice's weird modeling of the 555 and not how an actual 555 would behave, I'll try a real one later I guess. You could hold the trigger signal low or discharge the cap as alternate ways to stop the timer.
|
#
?
Feb 9, 2017 17:13
|
|
|
|
| # ? May 31, 2024 17:54 |
|
|
ate all the Oreos posted:I simulated pulling the Reset pin down in LTSpice and it didn't turn the oscillator off, it just made it so that the second it went high it went low again (so like a 1% duty cycle). Admittedly this is probably just LTSpice's weird modeling of the 555 and not how an actual 555 would behave, I'll try a real one later I guess. Yeah I know the 555 gets used a lot of different ways but for example the flip flip inputs could be yanked one way or another and should hold the output to one state.
|
|
#
?
Feb 9, 2017 17:16
|
|
|
taqueso posted:You could hold the trigger signal low or discharge the cap as alternate ways to stop the timer. Doesn't that make the output go high and stay there then? (That's what it did in LTSpice  ) I guess I could just invert the output in that case, hm.
|
|
#
?
Feb 9, 2017 17:35
|
|
|
Okay I just tried it IRL and pulling the reset pin to ground seems to just shut the whole thing off entirely so that'll work fine, thanks!
|
|
#
?
Feb 9, 2017 18:20
|
|
|
His Divine Shadow posted:I was just wondering if it might be feasible to make my homemade TIG pedal into a wireless with some simple off the shelf components, it uses a 10k potentiometer, don't know about precision, I just bought a cheap one and it works well enough, think it's a little more than one turn for it's range. Caveat: I know nothing about welding equipment. If the potentiometer is acting as a voltage divider and the (welding variable) you're adjusting is voltage controlled, you could probably just run in a PWM signal and get the same result, and there are far more and far cheaper options for that than actually automating the pot.
|
|
#
?
Feb 9, 2017 23:47
|
|
|
Spent the whole day reading up on how boost/buck converters work... I was wondering about a particular solar MPPT charge controller I was looking at: https://www.emarineinc.com/AERL-RACEMAX-MAXIMIZER-MPPT-Solar-Car-Controller-600B  Does anyone have any insight as to why this solar charge controller costs $1200? I mean, it is meant for solar race cars, so I guess the components are very high precision or something, and it claims a peak efficiency of 99%, but it seems like the whole thing is simply a DC-DC Boost converter that can taper a Li-Ion charge, and the output voltage can be adjusted by potentiometer. I'm not doubting there's a good reason it costs what it does, but is this something I could get a schematic and BOM for and replicate for a lot less money? Also, who's weiner do I have to lick to get one of these?: https://www.researchgate.net/publication/281232089_Full_SiC_DCDC-Converter_with_a_Power_Density_of_more_than_100_kWdm3  Yeah, that's a 200 kilowatt, 98% efficient liquid-cooled DC-DC converter that is 1.4 liters in volume. Does Mouser or anyone else sell those SiC mosfets?
|
|
#
?
Feb 10, 2017 07:11
|
|
|
poeticoddity posted:Caveat: I know nothing about welding equipment. I hooked up a multimeter to it when working and it does seem the pot controls voltage and I think there's a control board that interprets that voltage change in order to change the welding amperage. I think this is over my head, for now. So I'll stick to rebuilding it as a wired pedal, moving some of the components out of the pedal and mounting on the welder instead.
|
|
#
?
Feb 10, 2017 07:21
|
|
|
asdf32 posted:Winding your own core is cool. If you don't want to digikey will have a good selection of inductors that should work for boost converters in typical ranges of frequency/voltage/current. Thanks - will check that place out. Zero VGS posted:Does anyone have any insight as to why this solar charge controller costs $1200? Probably because it sells in low volume, so they have to make back their R&D + overheads by building it into the price. One day Dacien at Electrodacus may build a higher voltage design. His controllers are nice, and are more reasonably priced (but currently only at 12/24v) http://electrodacus.com/
|
|
#
?
Feb 10, 2017 12:25
|
|
|
Zero VGS posted:
This is my guess. At the currents and temperatures involved, this is absolutely bonkers.
|
|
#
?
Feb 10, 2017 15:34
|
|
|
Seems like for efficiencies like that you're getting to the point where stuff like exact construction materials of the board or which component is next to which other component matters so it might be hard to build your own without an engineering team and fancy simulation software.
|
|
#
?
Feb 10, 2017 15:38
|
|
|
Zero VGS posted:Does anyone have any insight as to why this solar charge controller costs $1200? quote:Does Mouser or anyone else sell those SiC mosfets? digikey mouser
|
|
#
?
Feb 10, 2017 16:59
|
|
|
Zero VGS posted:Spent the whole day reading up on how boost/buck converters work... It's low quantity. Building it probably costs a couple hundred and the R&D is significant particularly when you include the MPPT algorithm. Like most thing could could do it cheaper if you don't value your time.. SiC, like GAN, has sort of been 'around the corner' for a while. There are some parts available but not a lot. From what I've seen SiC have terrible thermal coefficients meaning they may dissipate somewhat less wattage but the junction will get hotter doing it. So you may actually need to bigger parts and/or heatsinks. That is fascinating. My task at the moment is a ~1.5kW inverter in about 0.65L and it's hard.. What's interesting is that 200khz isn't actually that high and they're not trying to soft switch, which is what you see in most similar papers (like the google challenge). Of course there is a little gamesmanship here - water cooling just externalizes the cooling volume. But 200KW is a huge number and that's a very small size.
|
|
#
?
Feb 10, 2017 17:06
|
|
|
A few newbie questions about EMI: If I can afford it, should I just go ahead and use shielded wire on everything? Do I need to ground the aluminum wrap to something? If so, if it okay if it all grounds to a metal table I'm working on?
|
|
#
?
Feb 11, 2017 04:03
|
|
|
CarForumPoster posted:A few newbie questions about EMI: I wouldn't bother, outside of analog signals in a very noisy environment (like audio when you hear stuff). It's slightly harder to work with/bulkier. Grounding the wrap improves it's shielding immensely. It'd be fine to connect it to the table.
|
|
#
?
Feb 11, 2017 04:22
|
|
|
Aurium posted:I wouldn't bother, outside of analog signals in a very noisy environment (like audio when you hear stuff). It's slightly harder to work with/bulkier. In this case Ill be transmitting digital, receiving digital, doing computer things and possibly receiving analog audio. Its a test bed so all of these things are fairly closeish together (most <6" apart) and theres a possibility of cosite issues.
|
|
#
?
Feb 11, 2017 04:31
|
|
|
CarForumPoster posted:In this case Ill be transmitting digital, receiving digital, doing computer things and possibly receiving analog audio. Its a test bed so all of these things are fairly closeish together and theres a possibility of cosite issues. I can't tell you it definitely won't be an issue. Digital typically doesn't need shielding until you get long runs or high speeds. It's also typically low current and doesn't generate much noise. It won't hurt, it will add fiddlyness. My suspicion is that you won't need it, I can't really give any other pointers.
|
|
#
?
Feb 11, 2017 04:52
|
|
|
Aurium posted:I can't tell you it definitely won't be an issue. Appreciated. Is there some neat trick to grounding the wire wrap? Whats the conventional method of grounding the wrap? Just tape it to the table? Alligator clips? Electrically bond my enclosures to the table and ground the wrap to the housings? Sorry for all the questions, suffice it to say its reasonable to expect this as an issue given the test requirements expected.
|
|
#
?
Feb 11, 2017 05:27
|
|
|
CarForumPoster posted:Appreciated. Is there some neat trick to grounding the wire wrap? Whats the conventional method of grounding the wrap? Just tape it to the table? Alligator clips? Electrically bond my enclosures to the table and ground the wrap to the housings? In many cables there's also a bare wire woven into the foil that you can solder to. Here's an example that someone in the CNC thread linked earlier. This is the easiest best way to do it. Many connectors crimp to shielding in addition to being connected to the internal wires. This is nice for production, not so much for prototyping unless you already know tons of connectors in all their sizes, as well as having the appropriate crimpers. After that, your other ideas would be fine.
|
|
#
?
Feb 11, 2017 05:49
|
|
|
CarForumPoster posted:Appreciated. Is there some neat trick to grounding the wire wrap? Whats the conventional method of grounding the wrap? Just tape it to the table? Alligator clips? Electrically bond my enclosures to the table and ground the wrap to the housings? You can buy solder sleeves with a lead-wire attached. Slip it over the insulation, hit it with a heat gun until the solder melts, and you've got a good ground connection. http://www.digikey.com/products/en/...d=0&pageSize=25
|
|
#
?
Feb 11, 2017 14:46
|
|
|
asdf32 posted:SiC, like GAN, has sort of been 'around the corner' for a while. There are some parts available but not a lot. From what I've seen SiC have terrible thermal coefficients meaning they may dissipate somewhat less wattage but the junction will get hotter doing it. I think silicon carbide devices can be engineered to tolerate higher junction temperatures when compared to silicon. People have done proof-of-principle demonstrations using silicon carbide for electronics in extreme environments, like > 200 C.
|
|
#
?
Feb 12, 2017 00:29
|
|
|
silence_kit posted:I think silicon carbide devices can be engineered to tolerate higher junction temperatures when compared to silicon. People have done proof-of-principle demonstrations using silicon carbide for electronics in extreme environments, like > 200 C. Interesting. The first one I pulled up was only 150C rated. Though they have a significantly smaller R increase with temperature which is nice (1.5X increase instead of 2-2.5X for regular fets)
|
|
#
?
Feb 12, 2017 00:43
|
|
|
silence_kit posted:I think silicon carbide devices can be engineered to tolerate higher junction temperatures when compared to silicon. People have done proof-of-principle demonstrations using silicon carbide for electronics in extreme environments, like > 200 C. There was just a pr for some NASA sic stuff that withstood simulated Venus temps Sic is great for high temps only diamond beats it
|
|
#
?
Feb 12, 2017 01:14
|
|
|
CarForumPoster posted:In this case Ill be transmitting digital, receiving digital, doing computer things and possibly receiving analog audio. Its a test bed so all of these things are fairly closeish together (most <6" apart) and theres a possibility of cosite issues. Are all the parts connected to the same power and ground? Also are you using single ended or differential signaling?
|
|
#
?
Feb 12, 2017 02:57
|
|
|
Keep in mind most of the hype surrounding SiC's high temperature performance ignores the fact that although the SiC itself maintains good semiconductor properties up to 500C, the gate oxide limits MOSFETs to maybe 200C. That's why all the super high temperature SiC circuits are based off of JFETs instead.Jamsta posted:Can anyone recommend a good place for buying ferrite cores. Aliexpress/Ebay sell cheap but they often just say the size and not the core grade. Is that going to be a problem? ANIME AKBAR fucked around with this message at 04:07 on Feb 12, 2017 |
|
#
?
Feb 12, 2017 03:46
|
|
|
Akbar do you have experience with SIC or Gan personally? I want to take a look at transform's cascode type GANs because I'll be able to drop them into a current design. TI pushes a much different solution with integrated gate drive.
|
|
#
?
Feb 12, 2017 16:17
|
|
|
asdf32 posted:Interesting. The first one I pulled up was only 150C rated. Though they have a significantly smaller R increase with temperature which is nice (1.5X increase instead of 2-2.5X for regular fets) To get the full benefit of the high temperature capability of silicon carbide, the device manufacturers have to make sure that the device metallization and the device packaging are all compatible with high temperatures. ANIME AKBAR posted:Keep in mind most of the hype surrounding SiC's high temperature performance ignores the fact that although the SiC itself maintains good semiconductor properties up to 500C, the gate oxide limits MOSFETs to maybe 200C. That's why all the super high temperature SiC circuits are based off of JFETs instead. I don't think that temperature has much to do with that. I think the reason why we don't often see silicon carbide, gallium nitride, gallium arsenide, germanium, or indium phosphide MISFET devices has to do with how the electronic surface property of silicon can be made to be so much better than that of the other semi-conductors. Silicon probably still holds the records for lowest surface recombination velocity & lowest interface trap density. Sufficiently low interface trap density is a technical requirement to make a good MISFET device. edit: This is wrong. There are silicon carbide MISFET/MOSFET devices. silence_kit fucked around with this message at 15:00 on Dec 1, 2019 |
|
#
?
Feb 12, 2017 19:16
|
|
|
So I think that I'm to the point that I need more kit. I want to put together a good selection of resistors and diodes and crap for basically teaching myself circuit schematics, tattooing KVL/KCL/Ohm's onto the insides of my eyelids, and teaching myself digital logic with actual physical parts. I know I need to accumulate a bag of 555 timers among other things and I've got a small selection of resistors and photodiodes and switches and a couple ICs and crap from my adafruit beginner's experiment kit however, I really don't know what to put together for a shopping list now. I already know plenty about software, and I'm trying to work backward to the point that I can put poo poo together to act like whatever I can do in a few lines of arduino code on a whim. I understand logic gates well enough but implementing them with physical parts is still something that I'm not really grasping, and I'd eventually like to be able to, say, build a binary block with raw electronic parts, etc, so I can either substitute something physical in lieu of a code-based solution, or build something the hard way and then cut it out and use easier tools. Before I can do that though, I suspect that I need to put together a decent selection of stuff that I don't even know what I need yet.. Mainly I just want to gently caress around with digital logic and do stuff like, say, making a small IC on a breadboard out of basic elements or something, so I can get a stronger internalization of how they work, etc. Does anyone have a link or know what I ought to start looking for/at? I'm still mainly working on small-scale stuff with hobby motors and LEDs and the like right now, although eventually I do plan on putting together some kind of generator as an Independent Study project for course credits, since I can't actually start formally taking electronics classes until Next Fall and by then if I can't teach half of it to myself, then I probably am out of my league. 
|
|
#
?
Feb 12, 2017 21:52
|
|
|
For breadboarding learning stuff, you need a fairly complete resistor starter kit, a less complete set of assorted caps, and a 7400 series logic set. At least a few ANDs, ORs, XORs, and inverters. Looked quickly and only saw crazy expensive $200 kits, but comb ebay or amazon for smaller kits, you don't really need that much.
|
|
#
?
Feb 13, 2017 20:13
|
|
|
You can buy resistors and capacitors in "kits" from amazon pretty cheap, here's the ones I got: 1200 resistors, 56 different values,  https://www.amazon.com/PIXNOR-Resistor-1ohm-10Mohm-Resistors-Assortment/dp/B00SD6F6FC/ 125 electrolytic caps, 25 values, https://www.amazon.com/Foxnovo-125pcs-Electrolytic-Capacitors-Assortment/dp/B00UJDJEAU/ 1000 ceramic caps, 50 values, 13 bux: https://www.amazon.com/Ltvystore-Capacitor-Assortment-1pf-100nf-Electronic/dp/B01H3H6KLU/ 30 different 7400 series logic chips, $20: https://www.amazon.com/Electronics-Salon-74HCxx-Assortment-High-Speed-Si-Gate/dp/B016ME1WH2/ NPN Transistors, 100 for $6: https://www.amazon.com/Uxcell-a11102000ux0360-Terminal-Through-Transistors/dp/B008IFYBWW You can get "kits" of random assortments of all sorts of different poo poo on amazon that people sell for slightly more than the cost of parting out each component separately. The difference is usually only a few dollars at most and it's way less hassle and most of them have same-day delivery if you have prime so ehhhh it's how i get most of my random jellybean parts. You can also look for "electronics grab bag" or similar, it's just random junk they had left over they throw in a bag and sell. You don't know what you're going to get but generally you'll at least get a bunch of different resistors, capacitors, LED's, diodes, etc. No guarantee that they'll be useful though
|
|
#
?
Feb 13, 2017 20:36
|
|
|
Those 1% resistors are notoriously not 1%, FWIW. You probably don’t need 1%, but still.
|
|
#
?
Feb 13, 2017 21:34
|
|
|
Platystemon posted:Those 1% resistors are notoriously not 1%, FWIW. I figured as much and tested a bunch and they measured what they said they did but ymmv 
|
|
#
?
Feb 13, 2017 21:41
|
|
|
ate all the Oreos posted:I figured as much and tested a bunch and they measured what they said they did but ymmv The fake resistor kits often are within 1%. The thing that makes them fake is that they're often not metal film, but the much cheaper carbon composition. For most hobby applications this doesn't matter, (other than being sold a lie) but they're a fair bit noisier, and thier temperature coefficient is far worse.
|
|
#
?
Feb 14, 2017 01:55
|
|
|
Aurium posted:In many cables there's also a bare wire woven into the foil that you can solder to. Here's an example that someone in the CNC thread linked earlier. This is the easiest best way to do it. Zuph posted:You can buy solder sleeves with a lead-wire attached. Slip it over the insulation, hit it with a heat gun until the solder melts, and you've got a good ground connection. You both rock. This is a huge area of concern for me based on reading past EMI test reports and it not being a thing I know much about. I'd rather it all be rock hard than deal with phantom issues during prototyping. Murgos posted:Are all the parts connected to the same power and ground? Also are you using single ended or differential signaling? Ground yes, power it'll mix. Most stuff is 5VDC and everything will trace back to the same batteries or power supply, but there are a few DC-DC converters in the planned mix. I am not sure to be honest, I am betting it will be a mix of both but I had to google what differential signaling was.
|
|
#
?
Feb 14, 2017 02:41
|
|
|
wow thanks a ton, that is EXACTLY the kind of responses I was hoping to receive! Off to go shopping now 
|
|
#
?
Feb 14, 2017 03:53
|
|
|
I have a variable voltage DC power supply with many end connections that I bought to remove the end to give a variety of breadboard voltages, by cutting the end of, and cutting the positive and negative wires to different lengths so they never touch. But thinking about it, if I just cut it off and put a plug on the end, and have a couple of sockets, I can have one socket connect to the positive/negative wires as before, but another socket that connects to the original ending so I can still use it for power plugs. What are the best connecting plug/sockets I can use for this purpose, if you follow my thoughts? Anything terrible bad about this idea? Just thinking I'd like to retain the usefulness of the original plug ending.
|
|
#
?
Feb 15, 2017 20:52
|
|
|
Fat Turkey posted:I have a variable voltage DC power supply with many end connections that I bought to remove the end to give a variety of breadboard voltages, by cutting the end of, and cutting the positive and negative wires to different lengths so they never touch. What kind of voltage/amperage are you talking? Because if it's not too much just use a barrel plug since they're super common and easy to work with e: Or wait do you mean you have a power supply that has multiple different wires coming out all at different voltages? I interpreted it as "I have a power supply I can change the voltage of" not "I have a power supply that outputs multiple voltages at once"
|
|
#
?
Feb 15, 2017 21:09
|
|
|
silence_kit posted:To get the full benefit of the high temperature capability of silicon carbide, the device manufacturers have to make sure that the device metallization and the device packaging are all compatible with high temperatures. We don't see exotic non silicon MOSFETs because silicon is dirt cheap to make and is pretty drat good until you need something extreme Rf has gaas and gan, you are seeing sic and gan in power MOSFETs slowly trickling in but price price price
|
|
#
?
Feb 15, 2017 21:27
|
|
|
Fat Turkey posted:I have a variable voltage DC power supply with many end connections that I bought to remove the end to give a variety of breadboard voltages, by cutting the end of, and cutting the positive and negative wires to different lengths so they never touch. Thanks to my quadcopter obsession, XT60 on everything is my favourite. Had to make converters to bare wire, breadboard, crocs, barrel, bullet, splitters and gender inverters. 
|
|
#
?
Feb 15, 2017 21:33
|
|
|
ate all the Oreos posted:What kind of voltage/amperage are you talking? Because if it's not too much just use a barrel plug since they're super common and easy to work with 3-12v, I have a 1 amp version and a 2 amp version. One cable, turn a dial on the plug for different voltages. Nothing major, just to power the prototypes of projects. The XT60s look a bit chunkier than I was expecting but are the kind of thing I mean. Anything a bit smaller? Fat Turkey fucked around with this message at 21:51 on Feb 15, 2017 |
|
#
?
Feb 15, 2017 21:44
|
|
|
|
| # ? May 31, 2024 17:54 |
|
|
Fat Turkey posted:3-12v, I have a 1 amp version and a 2 amp version. One cable, turn a dial on the plug for different voltages. Nothing major, just to power the prototypes of projects. 2.1mm barrel jacks would do, and are widely used. I've got a bunch with screw terminals on their ends for re-use. Personally don't like them as they fall out relatively easily. XT60's are fairly easy to disconnect but never fall apart, and obviously can handle a lot of current.
|
|
#
?
Feb 15, 2017 22:43
|
|