|
peepsalot posted:I'm interested in making my own capacitive discharge spot welder, for thin sheet metal, battery pack tabs, etc. I don't know much about welding but I see a number of issues here. First, as you calculated, your RC time constant is 28mS. How do you get multiple pulses out of it when it will be almost entirely discharged in 0.1 seconds? I can see why you'd use an SCR here. Also, if you actually did shut off this much current either intentionally or accidentally I could easily see the inductive spike blow this 40V fet. Second, given the short amount of time, paralleling is not a good idea. Mosfets are generally fine for paralleling because their resistance goes up with temperature, thus they self balance - but paralleling 10 isn't a great idea even in steady state. In this short amount of time however all the normal calculations go out the window. There is no chance for self balancing and consider what a small variation in turn on time would do - the first fet to come on would get roughly 10x the current you intend. So I don't recommend paralleling at all. However what's on your side is that the fets can handle vastly more burst current then they can steady state current. As an aside it's worth noting that the 195A you referenced in the datasheet is a largely useless number. In this case that's the package rating, meaning the TO-220 package and bond wire limit. The other number on the front page, 409A, is also a synthetic reference number that essentially assumes a perfect heatsink. So bottom line, neither really applies here. If you use the temperature numbers provided - 62 degrees rise per watt, 175C max junction temperature ~50C ambient temp, you'll see that this fet is only good for about 2W which gets ~40A continuous with no heat-sink. The pulsed drain current rating of ~1500A gives you another reference point. This is where I'm not an expert - I would guess that your pulse of ~50mS is wider than the type of pulse this is specifying. But I could be wrong. On your side is the fact that you're just doing one pulse. Note that because the time is so short, heat-sinking isn't going to matter unless somehow you're charging back up and doing another really quickly. Last you'd have to consider the drive circuity for this if you use an N-fet - you'd want to pull the gate above 12v to keep it on and ground bounce may be a big issue given the huge amounts of current. Power supplies often use isolation transformers on the gate control for this very reason (another reason for SCR's). If you mess around with this be careful. Caps that large are dangerous. What about 12V batteries? Those store tons more energy and can put out 100's of amps pretty easily. Another thing to remember is that the tab on these mosfets is usually electrically connected to the drain - that would be 12V in this case. If you can find any more references for this exact application I'd follow them - this is pretty specific. asdf32 fucked around with this message at 00:53 on Feb 25, 2013 |
# ? Feb 25, 2013 00:10 |
|
|
# ? May 9, 2024 02:29 |
|
We should really add some kind of "Don't solicit advice about high voltage/high amperage crap in the electronic hobby megathread on this (or any other) comedy forum." clause to the original post.
|
# ? Feb 25, 2013 00:18 |
What? Where do you think the comedy comes from?
|
|
# ? Feb 25, 2013 02:10 |
|
asdf32 posted:If you mess around with this be careful. Caps that large are dangerous. What about 12V batteries? Those store tons more energy and can put out 100's of amps pretty easily. Another thing to remember is that the tab on these mosfets is usually electrically connected to the drain - that would be 12V in this case. If you can find any more references for this exact application I'd follow them - this is pretty specific. What about an honest transformer-based spot welder? I have one of these for spot welding 20 thou stainless shim stock: http://www.harborfreight.com/230-volt-spot-welder-45690.html 20% off of that, and it works fine.
|
# ? Feb 25, 2013 02:26 |
MOSFETs are a very poor candidate for such high peak currents. Paralleling is difficult, not because the waveform is fast (28ms is a very long time, plenty long for a die to heat up quite a bit), but rather because they will tend to oscillate like crazy, since some of them will be operating in their triode region briefly. Here's a webpage from a guy who did roughly what you're proposing: http://99mpg.com/blog/makingacapacitordi The amount of work put into lowering ESR and ESL in his bus structure reminds me of what we did for our big solid state tesla coils. Lots of custom machined busbar. The FETs won't survive any significant amount of ESL, but SCRs put up with it fine (many even demand ESL to limit di/dt). And he also says he has do deal with gate oscillation (though he does so somewhat poorly...). I would only go for MOSFET switches if you absolutely want to do pulse width control or double pulse stuff. Otherwise, using a single rugged thyristor/SCRs is far simpler. I can pretty much guarantee the cost of all the custom busbar necessary for MOSFET system is greater than the cost of finding a few nice thyristors.
|
|
# ? Feb 25, 2013 14:48 |
|
I'm about to (probably) light $30 on fire here... http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=110925504614&ssPageName=ADME:X:RTQ:US:1123 I messaged the seller with the part number of a clean 15" 1680x1050 LCD out of a Latitude, and they say they can wire up a cable for it no problem. I've stripped a vintage-1998 lunchbox PC that was originally used as a Sprint network analyzer, and am shoving modern guts into it. This should work out for the display part nicely.
|
# ? Feb 25, 2013 20:13 |
|
If you need any sort of help with that sort of poo poo, I think there's a huge userbase of people that use those to make DIY cintiq clones over at http://forum.bongofish.co.uk/
|
# ? Feb 25, 2013 20:18 |
|
Anyone here have one of the Xilinx/Avnet Zedboards? We're putting a Zynq device down on an upcoming board design and I got one to prototype the software/hardware. It's really cool and a great exposure to the FPGA/ARM stuff if you're so inclined. It is a little pricey though at $395 ($319 educational price). Comes with the free webpack licenses and even a device locked chipscope pro.
|
# ? Feb 25, 2013 20:23 |
|
Jonny 290 posted:I'm about to (probably) light $30 on fire here... Please let us know how that works out. I'll have to check out the forum Martytoof linked, but I ordered one of those for a 17" Latitude panel and haven't been able to get it to work.
|
# ? Feb 25, 2013 22:00 |
|
priznat posted:Anyone here have one of the Xilinx/Avnet Zedboards? We're putting a Zynq device down on an upcoming board design and I got one to prototype the software/hardware. It's really cool and a great exposure to the FPGA/ARM stuff if you're so inclined. It is a little pricey though at $395 ($319 educational price). Comes with the free webpack licenses and even a device locked chipscope pro. So can you actually buy those Zynq chips yet? It seems like Xilinx has been hyping that poo poo for two or three years now, and every so often I check Avnet to verify that no, in fact you can't actually buy any chips.
|
# ? Feb 25, 2013 22:09 |
|
Poopernickel posted:So can you actually buy those Zynq chips yet? It seems like Xilinx has been hyping that poo poo for two or three years now, and every so often I check Avnet to verify that no, in fact you can't actually buy any chips. Yeah you can get em. The 7020 (one on the zedboard) and 7045 (kintex fabric with transceivers , $$$) are showing up now. Digikey has them, I'm sure those are being bought straight through Avnet (non-stock for digikey but parts available) so they probably have them too. The Xilinx rep I deal with says all of the parts plus 2 new ones they are about to announce will be available in production parts by the end of the year. They're really neat devices but unfortunately the development software is pretty lovely (no surprise to anyone who has ever used ISE). Apparently they will be rolling into the Vivado tools (still lovely but less so) "soon".
|
# ? Feb 25, 2013 22:35 |
|
asdf32 posted:I don't know much about welding but I see a number of issues here. First, as you calculated, your RC time constant is 28mS. How do you get multiple pulses out of it when it will be almost entirely discharged in 0.1 seconds? quote:I can see why you'd use an SCR here. Also, if you actually did shut off this much current either intentionally or accidentally I could easily see the inductive spike blow this 40V fet. quote:Second, given the short amount of time, paralleling is not a good idea. Mosfets are generally fine for paralleling because their resistance goes up with temperature, thus they self balance - but paralleling 10 isn't a great idea even in steady state. In this short amount of time however all the normal calculations go out the window. There is no chance for self balancing and consider what a small variation in turn on time would do - the first fet to come on would get roughly 10x the current you intend. quote:Last you'd have to consider the drive circuity for this if you use an N-fet - you'd want to pull the gate above 12v to keep it on and ground bounce may be a big issue given the huge amounts of current. Power supplies often use isolation transformers on the gate control for this very reason (another reason for SCR's).
|
# ? Feb 25, 2013 22:40 |
|
Delta-Wye posted:Talking about RF, is it just me or is a lot of RF/analog becoming a lost art? Seems like my parent's generation had quite a few self-taught HAM types who are comfortable with antenna design and a bunch of other things that I, with my formal education, am not comfortable with. It could just be practice (I don't do much antenna stuff, they have years and years of experience) or willingness to experiment (they are picking up analog radio stuff so they can play with the antenna, I'm doing digital 2.4 GHz so I follow the datasheet exactly), but I'm always sad when I see old electronics magazines and they gloss over details on stuff like antennas because 'duh' and I don't get it. Antenna design, on the other hand, has become easier to do and more popular than ever due to really powerful modelling software being either free or relatively cheap (<$100) and useful analyzers being sold for less than $400 (for the top-of-the-line model, others are cheaper).
|
# ? Feb 25, 2013 23:26 |
|
My wife and I are building a grow room for mushrooms (gourmet, not ) in our basement and I need something to help control the humidity. We already have a 120V plug-in pond fogger to supply the humidity. Well, the fogger's power supply is 120V. The fogger's element works at 24VAC at 8.33A and attaches to the power supply with some plug I've never seen before. Ideally I'd like to keep line voltage out of the grow room since it will be at around 80% humidity. Everything is already on GFCIs. The fogger element is on a 10 foot cord, so I can keep its power supply outside of the grow room. Well, I need some way of controlling the humidity. I was thinking of just mounting a regular 24VAC humidistat on the wall in there and connecting it in-line with the fogger element, but there are some problems with that: 1. 24V humidistats are meant for tiny solenoid valves for whole house humidifiers. They aren't rated for 8.33A 2. I've never seen the plug for the power supply before, and I don't want to go hacking up the wire just yet. The plug is some male bi-pin plug with 2mm pins, 6mm apart and 9mm long that has a plastic nut behind it that screws onto the power supply to hold it in. The only other option I could think of would be to control the power supply's input directly via a 24VAC transformer/relay combo with humidistat, but using a 24VAC transformer to turn on and off another 24VAC transformer seems a little silly. By the way, the only retail solutions I've seen for this wouldn't work. They either plug in directly and have a receptacle on their face (I'm not putting an outlet and the power supply in there), or they cost over $150 and I know I could build something cheaper than that. kid sinister fucked around with this message at 19:28 on Feb 26, 2013 |
# ? Feb 26, 2013 01:24 |
|
I know you've been getting advice on timers in the hydroponics thread, but if you're willing to build your own solution you could probably get a feedback system going pretty cheaply. You'll need a humidity sensor that outputs a simple variable voltage signal (something like a HIH-4030 would be really easy to work with), a mosfet or relay for switching on the fogger, and a comparator or arduino-like microcontroller, plus a potentiometer if you're using the comparator instead of a programmable controller. If you're using a comparator, hook up the humidity sensor signal wire to the + input and a voltage source of ~5v via the trimpot to the - input. You want to adjust the trimpot so that the voltage going to the - input of the comparator is around the same voltage level the humidity sensor produces at your target RH% (approx 3.3v at 80% for the one linked). The output of the comparator will then go high when the humidity drops below 80% and go low when it goes above 80%. You can then feed this to the relay (may need a boost circuit to allow enough current through to close the relay) which you'd need to splice into the power cable going to the fogger. Using an arduino you'd read the signal, when it goes below your threshold voltage switch on an output to close the relay and switch on the fogger. When the signal goes above the threshold switch the output off again. You might even be able to use a pair of mosfets to switch the AC circuit but I'm not sure how well that would work.
|
# ? Feb 27, 2013 06:13 |
|
Ephphatha posted:You might even be able to use a pair of mosfets to switch the AC circuit but I'm not sure how well that would work.
|
# ? Feb 27, 2013 06:36 |
|
kid sinister posted:Well, I need some way of controlling the humidity. I was thinking of just mounting a regular 24VAC humidistat on the wall in there and connecting it in-line with the fogger element, but there are some problems with that: My friend got this sensor from sparkfun and made a humidor. I2C, cheap, accurate. It was so fast and easy (with a PIC!), he got bored Saturday after making the thing, then bought a combo temp/humidity, a 16-line LED display with tricolor backlight, some rotary encoders, and an ethernet module. Getting all THAT crap hacked together took like four weekends. If you don't like that sensor, try this one and a comparator hooked to a relay.
|
# ? Feb 27, 2013 11:17 |
|
babyeatingpsychopath posted:My friend got this sensor from sparkfun and made a humidor. I2C, cheap, accurate. It was so fast and easy (with a PIC!), he got bored Saturday after making the thing, then bought a combo temp/humidity, a 16-line LED display with tricolor backlight, some rotary encoders, and an ethernet module. Getting all THAT crap hacked together took like four weekends. Make a network-connected thing like this guy did and create a mushroom farm that tweets you status updates on the mushrooms, because that would be adorable
|
# ? Feb 27, 2013 19:08 |
|
You guys are trying to reinvent the wheel. I would just use a mechanical humidistat. All they are is an SPST switch that closes when the humidity falls below the dial setting. In fact I wonder if I might be able to get away with a 120v humidistat if they support higher amperages since they're just switches...
|
# ? Feb 27, 2013 21:30 |
|
kid sinister posted:You guys are trying to reinvent the wheel. I would just use a mechanical humidistat. All they are is an SPST switch that closes when the humidity falls below the dial setting. In fact I wonder if I might be able to get away with a 120v humidistat if they support higher amperages since they're just switches... This is the learning electronics thread, not the granpda thread. ya grandpa.
|
# ? Feb 27, 2013 21:37 |
|
kid sinister posted:You guys are trying to reinvent the wheel. I would just use a mechanical humidistat. All they are is an SPST switch that closes when the humidity falls below the dial setting. In fact I wonder if I might be able to get away with a 120v humidistat if they support higher amperages since they're just switches... You're the one that doesn't want to hack up a cord. Cut that thing and put an inline in it.
|
# ? Feb 27, 2013 21:58 |
|
babyeatingpsychopath posted:You're the one that doesn't want to hack up a cord. Cut that thing and put an inline in it. He also didn't want 120V in the room. I have no idea how a fully electromechanical humidistat would work, but presumably it would require a certain voltage range to work. All that I have seen do all the sensing work with a sensor, a microcontroller, and a relay at low voltages. Essentially, they do exactly what people are proposing here.
|
# ? Feb 27, 2013 22:07 |
|
Slanderer posted:He also didn't want 120V in the room. I have no idea how a fully electromechanical humidistat would work, but presumably it would require a certain voltage range to work. All that I have seen do all the sensing work with a sensor, a microcontroller, and a relay at low voltages. Essentially, they do exactly what people are proposing here. Having read all his posts everywhere about this, I still maintain that 80% humidity is just fine for all kinds of stuff. Even consumer-grade crap stuff with the UL stamp on it is rated 5-95% humidity non-condensing at 0-130F. Reinventing the wheel is inventing complicated arrangements of parts to keep products out of an environment within the specs of the product. If any of his parts work inside a garage in Georgia in the summer, they'll be fine in this grow room.
|
# ? Feb 27, 2013 22:23 |
|
Slanderer posted:This is the learning electronics thread, not the granpda thread. = me All I was saying is that there's already an off-the-shelf part to do what you want to do, instead of hacking together a humidity sensor, microcontroller and a pot. babyeatingpsychopath posted:You're the one that doesn't want to hack up a cord. Cut that thing and put an inline in it. Find me an inline solution that could take 8.33A and I'll consider it. Now that I think about it, that bi-pin connector with a screw-on backer nut kind of sounds like a Mil-Spec connector. Is anyone here familiar with those? I'm not. babyeatingpsychopath posted:Having read all his posts everywhere about this, I still maintain that 80% humidity is just fine for all kinds of stuff. Even consumer-grade crap stuff with the UL stamp on it is rated 5-95% humidity non-condensing at 0-130F. Reinventing the wheel is inventing complicated arrangements of parts to keep products out of an environment within the specs of the product. If any of his parts work inside a garage in Georgia in the summer, they'll be fine in this grow room. You wouldn't need to add humidity, a Georgian garage in the summer is already at 80% Ugh, nevermind, apparently the point is moot. According to mushroom grower forums, mushroom spores tend to gently caress up humidity sensors within just a few hours of exposure. Back to the drawing board... kid sinister fucked around with this message at 23:35 on Feb 27, 2013 |
# ? Feb 27, 2013 22:41 |
|
We're tearing apart an electronic semi-automatic Nerf gun for a school project. I've created a circuit diagram from the circuit board which is controlling two motors which rotate in opposite directions (added some labels in paint that I forgot): Here is a picture of the circuit board and motors (inductors are on the back, battery pack and switch are elsewhere): We're hoping to identify the point of the capacitors and inductors for our report. My guess for the inductors are that when you depress the trigger (close the switch) there is a huge spike in current as the motors spin up. Since the current through an inductor cannot change instantaneously, they help to smooth out/slow down that spike in current. The capacitors have us stumped. The G1 and G2 nodes are soldered to the outer casing of the motors. Anyone have any ideas?
|
# ? Feb 28, 2013 01:59 |
|
alwayslost posted:We're tearing apart an electronic semi-automatic Nerf gun for a school project. I've created a circuit diagram from the circuit board which is controlling two motors which rotate in opposite directions (added some labels in paint that I forgot): What kind of Nerf gun is it? I'd just like to know because I have one of the electric ones and if we have the same one it would be neat to know about the motor circuitry I am a huge manchild
|
# ? Feb 28, 2013 02:03 |
|
Nerf N-Strike Elite Hail Fire
|
# ? Feb 28, 2013 02:26 |
|
alwayslost posted:Nerf N-Strike Elite Hail Fire Oh, that's the fancier one, I have the cheap pistol "Barricade" thing. Anyway, the motors have load applied and removed quickly (when they fire a shot), could the capacitors be there to help handle the sudden temporary massive increase in load?
|
# ? Feb 28, 2013 03:03 |
alwayslost posted:We're tearing apart an electronic semi-automatic Nerf gun for a school project. I've created a circuit diagram from the circuit board which is controlling two motors which rotate in opposite directions (added some labels in paint that I forgot): But C7 is definitely not how you drew it in the schematic. There's no way the motors would work with a capacitor in series. Are you sure it's not in parallel with the input leads?
|
|
# ? Feb 28, 2013 05:36 |
|
ANIME AKBAR posted:It's unlikely the inductors are big enough to significantly reduce inrush current on startup. The caps and inductors are likely just there to filter out the lovely noise generated from cheap brushed DC motors. Why would you need to filter noise if it's basically just a battery, motors and a switch? Will the noise gently caress with the battery or something?
|
# ? Feb 28, 2013 06:09 |
|
ANIME AKBAR posted:It's unlikely the inductors are big enough to significantly reduce inrush current on startup. The caps and inductors are likely just there to filter out the lovely noise generated from cheap brushed DC motors. Filtering really doesn't make sense in this case, since the whole thing is battery powered and without active electronics. I'm more curious about G1 and G2, frankly. C7 does look wrong, unless the circuit was exceptionally weirder than it currently is. EDIT: The circuit was to be doing something to turn the motors on for some amount of time and then stopping, but I'm honestly stumped. Slanderer fucked around with this message at 07:02 on Feb 28, 2013 |
# ? Feb 28, 2013 07:00 |
|
I can't speak to this myself for sure, but one of my friends recently disassembled an automatic Nerf gun and was very confused about a cap in the same place as that C7. I think what he eventually found out was that it's a diode in a package that looks a lot like a cap (and possibly mislabeled?), which makes more sense as it would prevent anything from happening if you put the batteries in the wrong way.
|
# ? Feb 28, 2013 07:10 |
|
We were confused by C7 too. If its just a diode that makes a lot more sense.
|
# ? Mar 1, 2013 03:29 |
|
I gave in and subscribed to Every freakin industry rag today. Give me your thinly veiled ads intermixed with normal ads, give me all of them unf
|
# ? Mar 1, 2013 16:01 |
|
Otto Skorzeny posted:I gave in and subscribed to Every freakin industry rag today. Give me your thinly veiled ads intermixed with normal ads, give me all of them unf Heh, it's true but don't discount the ads. I've had to re-program my reading habits when reading industry mags because the ads are important! If a company is advertising something it means 1) they think it's good and I might not know about it 2) the product availability is probably good 3) they're probably going to produce it for a while. All of those are good criteria for designing something in. I've found more useful components from ads than from articles. Ads in an industry mag arn't like pepsi fighting coke for image - with 500k+ parts at digi-key alone I have no doubt that there are a ton of components out there that I don't know about but should know about. Ads a legit way to find them.
|
# ? Mar 2, 2013 23:54 |
|
So I picked up a Motorola Lapdock for $50 to use with my Raspberry Pi. Instead of waiting for a micro-HDMI adapter to arrive from China, I've decided to just mod the unit. I want to cut the existing micro HDMI male connector off and hardwire an HDMI cable straight in. I'm perfectly capable of doing this but I have a question regarding RF shielding. Would wrapping some tin foil around the new wire bundle be sufficient for shielding or will I need some sort of metallic tape? Here's an example of the bare connector I'm going to be snipping off: http://andreiprojects.blogspot.ro/2012/06/raspberry-pi-modding-atrix-lapdock.html
|
# ? Mar 3, 2013 07:48 |
|
HATE TROLL TIM posted:So I picked up a Motorola Lapdock for $50 to use with my Raspberry Pi. Instead of waiting for a micro-HDMI adapter to arrive from China, I've decided to just mod the unit. I want to cut the existing micro HDMI male connector off and hardwire an HDMI cable straight in. I'm perfectly capable of doing this but I have a question regarding RF shielding. If you're just removing an inch or two of shield it's no big deal. Just leave it with no shield.
|
# ? Mar 3, 2013 16:49 |
|
Okay, I'll see how that works. I was under the impression HDMI is very susceptible to noise. If I *did* need shielding, would tin foil do the job?
|
# ? Mar 4, 2013 03:07 |
|
HATE TROLL TIM posted:Okay, I'll see how that works. I was under the impression HDMI is very susceptible to noise. If I *did* need shielding, would tin foil do the job? Consider that they sell 25'+ cables so there is a good amount of margin built into the standard. If you've got 6' and an inch unsheilded it's going to be fine. Also consider that the PCB traces aren't shielded, impedance controlled, but not shielded. Also it's a digital bus so working or not working will be fairly black and white, unlike old analog TV with static etc. But yeah I think tin foil would be ok, copper tape would be easier. Ideally you'd connect the shield to a "drain", probably the board ground would be fine - see what the data pair shields are going to on the board.
|
# ? Mar 4, 2013 03:40 |
|
|
# ? May 9, 2024 02:29 |
|
Just a quick-ish audio question - I'm playing with sound on my STM32F0DISCOVERY, and so I don't end up wrecking anything I'm running the output through an LM324 op-amp. I'm running it at 5 volts (well, the onboard regulator is giving me 4.6 volts but I'll take it I guess). It's reasonably loud in headphones, really quiet on an 8 ohm PC speaker I have. Most signs point to using an LM386 instead because it's specifically meant for audio output. Looks like it can source about twice the current, though the LM386 datasheet lists output power (in watts) and the LM324 lists output current, so that's a little confusing. Anyway, my question is: would the LM324 be able to get louder if I ran it with a higher voltage? The LM386 datasheet shows higher output power as voltage increases. Just trying to make sure I'm learning electronics the right way.
|
# ? Mar 4, 2013 06:37 |